#ceph IRC Log

Index

IRC Log for 2013-09-05

Timestamps are in GMT/BST.

[0:00] <sagewk> and there should be an opposite option too, so that the default can be modified in distro packages but users can explicitly ask for a particular behavior
[0:00] <alfredodeza> oh?
[0:00] <alfredodeza> even if the default *is* the opposite ?
[0:00] <vanham> sagewk, my ceph status is showing HEALTH_WARN but it doesn't say why. Now all OSDs, MDSs and MONs are up and running
[0:00] <sagewk> the ubuntu package will make --preserve-repos the default
[0:00] <vanham> any tip?
[0:01] <sagewk> vanham: ceph health detail
[0:01] * alfredodeza realizes this needs even more thought
[0:01] <sagewk> --preserve-repos vs --adjust-repos?
[0:01] <sagewk> or --adjust-repos vs --no-adjust-repos
[0:02] <alfredodeza> *that*
[0:02] <sagewk> \o/
[0:02] <alfredodeza> bravo
[0:02] <kraken> http://i.imgur.com/wSvsV.gif
[0:02] <alfredodeza> norris sagewk
[0:02] <kraken> sagewk remembers the future.
[0:02] <sagewk> if only
[0:02] <dmick> --preserve-adjustment-to-repos
[0:03] <vanham> sagewk, free space, thanks
[0:03] <dmick> :P
[0:04] <vanham> sagewk, to close the day, my MDS was stuck at rejoin, getting blacklisted. I already formatted it but, what should I have done?
[0:05] <zackc> uhaha
[0:05] * zackc likes --preserve-repos and --mangle-repos
[0:06] <alfredodeza> zackc: the negation is what kills it IMO
[0:06] <alfredodeza> if the flag is ON/OFF the negation has to be very similar to the positive (except for, negating it)
[0:06] <vanham> sagewk, when I added a higher debug level for the mds and ms, it was looking for some inode number, in some kind of loop
[0:08] <xarses> alfredodeza: love the new flag
[0:08] <xarses> dont care about the name of it =)
[0:09] <sagewk> vanham: post the logs so we could look..
[0:10] <sagewk> alfredodeza: will --no-adjust-repos *remove* previously adjusted repo files?
[0:10] <sagewk> --adjust-repos / --restore-repos / --no-adjust-repos :)
[0:11] <alphe> sagewk the .rgw as to ne big or the .rgw.bucket ? I dont understand in radosgw where really my data are stored in what pool
[0:11] <sagewk> can't win!
[0:11] <sagewk> .rgw.data i think it's called?
[0:12] <alphe> in the list of pools to create in that page http://ceph.com/docs/master/radosgw/config/
[0:12] <alphe> there is no mention of a .rgw.data
[0:12] <alphe> no mention of a rgw.bucket neither though ...
[0:13] <vanham> sagewk, too many lines for the whole thing but: 10 mds.0.cache _open_ino_backtrace_fetched ino 1000002f7fc errno -2 / 10 mds.0.cache no object in pool -1, retrying pool 1 / 1 -- 10.3.0.1:6800/19712 --> 10.3.1.2:6800/25090 -- osd_op(mds.0.48:8473969 1000002f7fc.00000000 [getxattr parent] 1.7043db5e e8887) v4 -- ?+0 0x7847200 con 0x28b2b00
[0:13] <alphe> radosgw is supposed to create the pools automatically when you run it for first time according to that same doc page
[0:14] <vanham> sagewk, whole log is 24 million lines long
[0:14] <sagewk> vanham: oh, that sounds like the bug alexandre oliva just fixed..
[0:14] <vanham> nice :)
[0:14] <vanham> going for a 0.67.3?
[0:14] <sagewk> yeah
[0:14] <vanham> Okay, thanks man!
[0:14] <alphe> mds can be blacklisted ?
[0:14] <sagewk> vanham: already in the dumpling branch
[0:14] <alphe> rho ...
[0:14] <alphe> poor mds ...
[0:15] <vanham> alphe, yeah...
[0:15] <sagewk> sjusthm: have a minute to talk about repgather vs new message?
[0:15] <alphe> how do you see concretely that your mds is blacklisted ?
[0:15] <alphe> because it always apears offline ?
[0:15] <sagewk> alphe: ceph osd dump
[0:16] <vanham> sagewk, thank you sage! I really like Ceph as a whole, with RBD and everything, CephFS has a lot of value for me
[0:16] <sagewk> vanham: :)
[0:16] <dmsimard> Another question on the topic of zones and topology. Is there a way to manage the topology other than manually editing the CRUSH map ?
[0:16] <vanham> sagewk, using it with Active, StandBy
[0:16] <sjusthm> sagewk: yep
[0:16] <sjusthm> core room?
[0:16] <sagewk> sure
[0:16] <vanham> Heading out,
[0:17] <vanham> sagewk, thanks and have a great one!
[0:17] * vanham (~vanham@187.115.200.154) Quit (Quit: Saindo)
[0:22] * rturk-away is now known as rturk
[0:24] <dmsimard> Managing the topology manually through the CRUSH map seems easy enough, nevermind my question :D
[0:25] * AfC (~andrew@101.119.15.156) has joined #ceph
[0:26] * dmsimard (~Adium@108.163.152.2) Quit (Quit: Leaving.)
[0:32] <alfredodeza> sagewk: no, the negation will not remove anything, it will just skip the mangling
[0:39] * BManojlovic (~steki@fo-d-130.180.254.37.targo.rs) Quit (Quit: Ja odoh a vi sta 'ocete...)
[0:44] * jks (~jks@3e6b5724.rev.stofanet.dk) Quit (Read error: Connection reset by peer)
[0:49] * sleinen (~Adium@2001:620:0:25:48d4:6aa9:7ce9:790a) Quit (Quit: Leaving.)
[0:49] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[0:49] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) Quit (Quit: Leaving.)
[0:50] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) has joined #ceph
[0:50] * AfC (~andrew@101.119.15.156) Quit (Read error: No route to host)
[0:57] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[0:57] * diegows (~diegows@190.190.11.42) Quit (Ping timeout: 480 seconds)
[0:59] * asomers (~alans@207-225-98-3.dia.static.qwest.net) has joined #ceph
[1:00] <asomers> Could anybody give me some advice about getting started developing ceph?
[1:00] <asomers> My first priority will be working on the FreeBSD port.
[1:01] <asomers> Where I most need advice is on testing. What regression tests should I run to validate my work?
[1:03] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) Quit (Quit: ...)
[1:05] <joshd> asomers: a good place to start with tests is the ceph_test_* binaries. there are also scripts in qa/workunits/* to run them and test other things like cli tools
[1:05] <sjusthm> asomers: ultimately, you'll want to try to adapt teuthology and the associated suites
[1:06] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) has joined #ceph
[1:06] <joshd> asomers: for freebsd, you'll probably be interested in http://wiki.ceph.com/01Planning/02Blueprints/Emperor/Increasing_Ceph_portability
[1:07] <asomers> So I should run all of the ceph_test_* and expect 0 exit status?
[1:07] <asomers> I can't find any "qa" directory in my ceph source directory. Is that a separate download?
[1:07] * Cube (~Cube@12.248.40.138) has joined #ceph
[1:08] <joshd> it's in the root dir, above src
[1:08] <joshd> some of the ceph_test_* binaries are more complicated, and would need e.g. radosgw set up, or extra parameters
[1:08] <asomers> I don't see it. Is that only in the git checkout? I downloaded a release tarball.
[1:09] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) Quit (Ping timeout: 480 seconds)
[1:09] <joshd> it's in git, maybe not the tarball
[1:09] <asomers> Yep. I see it on github now.
[1:09] <joshd> ceph_test_rados* is a nice subset that doesn't need any special setup, just look for 0 exit status
[1:10] <asomers> awesome. It uses googletest.
[1:11] <joshd> these all require a running cluster, and probably use client.admin (the default user) and the default search path for finding ceph.conf
[1:11] <alphe> ircolle I found what caused my ceph nodes to not be able to send data while they can receive data on the network and that was the cause of the client side ceph-fuse freeze I get. It is due to the fact that the default gateway route was sometimes set as the cluster private replication lan instead of the public lan gateway so you could connect from the public to it be it could not connect to the public
[1:14] <asomers> Are there any unit tests that don't require running servers?
[1:14] <alphe> ircolle to solve that issue I simply removed the option routers XX.XX.XX.XX from the dhcpd server of my cluster private lan
[1:14] <ircolle> alphe - thanks for the update
[1:15] <joshd> asomers: a few, but not nearly as comprehensive - there are some unittest_* binaries and other things like that run by 'make check'
[1:15] <alphe> ircolle you are welcome I am not sure what I said had sense to you :)
[1:16] <alphe> but think of it as a default gateway race problem ...
[1:17] <asomers> Thanks for the tips. It seems that my strategy should be:
[1:17] <asomers> 1) Write the port
[1:17] <asomers> 2) Run unit tests with "make check"
[1:17] <asomers> 3) Identify which of qa/workunits and ceph_test_* require no special setup
[1:17] <asomers> 4) Wrap the above in ATF
[1:17] <asomers> 5) Cross fingers
[1:19] * portante is now known as portante|afk
[1:19] <alphe> when I have mds cluster degraded what can i do ?
[1:19] <alphe> health HEALTH_WARN 158 pgs stale; 158 pgs stuck stale; mds cluster is degraded; clock skew detected on mon.osd03
[1:20] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[1:21] <joshd> asomer: sounds like a good start
[1:23] * asomers (~alans@207-225-98-3.dia.static.qwest.net) has left #ceph
[1:25] * MACscr (~Adium@c-98-214-103-147.hsd1.il.comcast.net) Quit (Ping timeout: 480 seconds)
[1:26] <xarses> alphe, well from that message, it sounds like your time isn't in sync
[1:26] <xarses> check your ntpd process
[1:27] * nwat (~nwat@eduroam-237-79.ucsc.edu) Quit (Ping timeout: 480 seconds)
[1:27] <alphe> yep
[1:28] <alphe> for a mysterious reason my main ntpd stoped to give the time
[1:28] <xarses> your env's ntp server stopped? or the client ntpd?
[1:29] <xarses> how many ntp sources do you have defined?
[1:29] <alphe> the clients can t connect to my local ntpd ...
[1:30] <alphe> probably a problem with the broadcast setting of my ntpd relay
[1:30] <xarses> the ntp service (at least for RHEL land) dosn't stay in sync properly if there are fewer than three sources
[1:30] <alphe> hum in ubuntu
[1:31] * alfredodeza (~alfredode@c-24-131-46-23.hsd1.ga.comcast.net) Quit (Remote host closed the connection)
[1:31] <xarses> i haven't worked with ntpd enough in ubuntu, so i cant say how the newer version reacts
[1:32] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) has joined #ceph
[1:32] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[1:33] <xarses> but the RHEL versions need three sources, otherwise the skew will reach the panic interval quite quickly (hours to a couple of weeks) in a virtual env (depending on hypervisor load) and once it reaches the panic interval it will just shutdown
[1:33] <xarses> its quite amusing, we had to check ntpd daily on our servers
[1:35] <alphe> yes ...
[1:36] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[1:39] * kraken (~kraken@c-24-131-46-23.hsd1.ga.comcast.net) Quit (Ping timeout: 480 seconds)
[1:42] <alphe> ok so the clock skew is gone only stays the mds degraded and the pg stuck
[1:44] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[1:44] * wschulze (~wschulze@cpe-69-203-80-81.nyc.res.rr.com) Quit (Quit: Leaving.)
[1:44] * grepory (~Adium@250.sub-70-197-12.myvzw.com) has joined #ceph
[1:47] * grepory (~Adium@250.sub-70-197-12.myvzw.com) Quit ()
[1:49] <alphe> osdmap e160: 20 osds: 5 up, 5 in
[1:50] <alphe> this is not good and all the nodes osd logs have error the ressource is busy
[1:51] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[1:51] * sherry (~sherry@wireless-nat-10.auckland.ac.nz) has joined #ceph
[1:51] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) has joined #ceph
[1:52] <alphe> ok osd are not getting up ...
[1:53] <xarses> alphe: ya, sounds like there aren't enough osd's up to satsfy the data placement requirements
[1:53] <sherry> what is snapid_t besides object_t used for?
[1:53] <alphe> ** ERROR: error converting sto
[1:53] <alphe> re /var/lib/ceph/osd/ceph-10: (16) Device or resource busy
[1:53] <alphe> ....
[1:53] <xarses> alphe: is /var/lib/ceph/osd/ceph-10 mounted?
[1:53] <alphe> i don t think so
[1:53] <xarses> is it a device?
[1:54] <xarses> or supposed to be a folder?
[1:54] <alphe> how can i know that ?
[1:54] <alphe> I ceph-deployed my osd so ...
[1:54] <xarses> is it supposed to point to /dev/sdb?
[1:54] <xarses> or something similar
[1:54] <sherry> what is snapid_t besides object_t used for?
[1:55] <alphe> i don t know ...
[1:55] <alphe> probably
[1:55] <xarses> ok, so if it's a device its supposed to be mounted
[1:55] <xarses> it should show up in df
[1:55] * ircolle (~Adium@c-67-165-237-235.hsd1.co.comcast.net) Quit (Quit: Leaving.)
[1:55] <alphe> yeah
[1:55] <alphe> and it doesn t show up
[1:56] <xarses> ok, but some of the others do?
[1:56] <xarses> can you post your df, and ceph osd tree?
[1:57] <alphe> ceph osd tree is jammed
[1:58] * malcolm (~malcolm@silico24.lnk.telstra.net) has joined #ceph
[1:58] <alphe> for example it shows me a ceph osd running in my 5th node I check that node and there is nothing
[1:58] <alphe> no disks mounted no ceph-osd in process list
[1:59] <alphe> ok think I will purge that thing now that the route probleme is solved
[1:59] <alphe> and reinstall the ceph cluster
[1:59] <xarses> you can service ceph -a start on the osd nodea
[2:00] <xarses> nodes even
[2:00] <xarses> they might start
[2:00] <xarses> but it sounds like something is accessing the device
[2:00] <xarses> which would explain the busy message
[2:00] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) Quit (Quit: sprachgenerator)
[2:01] <alphe> invalid command
[2:01] <xarses> otherwise the issue seems a bit more advanced than me
[2:01] <alphe> /etc/init.d/ceph -a start worked
[2:01] <alphe> only monitors are comming up
[2:01] <joshd> sherry: snapid_t is the snapshot of an object to read from, while object_t is just the name of the object
[2:02] <alphe> which is logical since the disks for the ceph-osd process are not mounted
[2:02] <malcolm> Hi all, just a quick one no rush on the answer/comments. I was testing CephFS and had some striping values set. I expected to see performance greater than one disk, was that an incorrect expctation? my stripe unit was quite long and my objects quite large. Count was 12 (24 disks in the cluster)
[2:02] <malcolm> Adn test was big dd's
[2:02] <xarses> alphe ceph -a start should start all the osd's that the host owns also
[2:02] <alphe> why the disks that the osd want are not mounted anymore ?
[2:03] <xarses> including mounting them
[2:03] <alphe> filestore(/var/lib/ceph/osd/ceph-0) l
[2:03] <alphe> ock_fsid failed to lock /var/lib/ceph/osd/ceph-0/fsid, is another ceph-osd still
[2:03] <alphe> running? (11) Resource temporarily unavailable
[2:03] <xarses> check the ceph log
[2:03] <alphe> I think that is the issue ...
[2:04] <xarses> ps axu | grep ceph-osd
[2:04] <alphe> they don t auto mount because they see the disk as already mounted since the fsid is still there
[2:04] <xarses> otherwise the device is readonly
[2:04] <alphe> ps aux | grep ceph-osd
[2:04] <xarses> which would also prevent the lock
[2:04] <alphe> root 1556 0.0 0.0 9676 956 pts/1 S+ 20:04 0:00 grep --color=auto ceph-osd
[2:05] <alphe> think I will have to remove those locks
[2:05] <xarses> ya, I'd try a reboot for funzee's to see if it clears it
[2:05] <xarses> or rebuild you cluster like you said
[2:05] <alphe> rebooted 3 times already and fsid files persists
[2:05] <xarses> odd
[2:05] * smiley_ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) has joined #ceph
[2:06] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[2:07] * alram (~alram@38.122.20.226) Quit (Quit: leaving)
[2:09] * portante|afk is now known as portante
[2:13] <alphe> I think I will crush that ceph cluster and reinstall it
[2:13] <alphe> osd are not showing logs
[2:13] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) Quit (Ping timeout: 480 seconds)
[2:14] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) has joined #ceph
[2:17] <xarses> if you dont re-image the nodes, make sure to use ceph-deploy purgedata and ceph-deploy purge
[2:22] <alphe> yep
[2:22] <alphe> what is re imaging ?
[2:22] * AfC (~andrew@2407:7800:200:1011:6e88:14ff:fe33:2a9c) has joined #ceph
[2:23] <xarses> re-install the host operating system
[2:23] <alphe> for some strange reason the ceph-osd crash directly and don t even write in logs
[2:23] <alphe> ho
[2:23] <joshd> yo61: it sounds like you might be hitting http://tracker.ceph.com/issues/3737, which is fixed by upgrading qemu to 1.4.2 or higher assuming you're running a a recent release of bobtail, cuttlefish, or dumpling
[2:23] <alphe> I will purge data purge uninsall and remove the partitions of my disks
[2:24] <xarses> ok, the "ceph-deploy disk zap" should nuke the partations nicely
[2:24] <alphe> ok
[2:24] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) Quit (Quit: Leaving.)
[2:24] <alphe> so I start with that then purge data purge
[2:24] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[2:25] <xarses> datapurge would run first
[2:25] <xarses> purgedata deven
[2:25] <xarses> it erases verything in /var/lib/ceph/*
[2:25] <alphe> how i disk zap it the client have no data ?
[2:26] <alphe> the ceph cluster is stoped anyway
[2:28] <xarses> the disk zap will erase the partition table on the disk and repartition it
[2:29] <xarses> purgedata wipes /var/lib/ceph/*
[2:29] <xarses> if anything was mounted, it would erase them too
[2:29] * angdraug (~angdraug@204.11.231.50.static.etheric.net) Quit (Quit: Leaving)
[2:29] * sagelap1 (~sage@2600:1012:b02d:5fe7:a1d8:6962:87a3:483b) has joined #ceph
[2:32] <alphe> yep
[2:33] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[2:33] <alphe> but nothing was mounted nowhere
[2:33] <alphe> so ...
[2:34] * sagelap (~sage@2607:f298:a:607:ea03:9aff:febc:4c23) Quit (Ping timeout: 480 seconds)
[2:40] * grepory (~Adium@2600:1010:b016:9fb4:bce3:6388:3bff:1ce4) has joined #ceph
[2:44] * xarses (~andreww@204.11.231.50.static.etheric.net) Quit (Read error: Operation timed out)
[2:49] * nerdtron (~kenneth@202.60.8.252) has joined #ceph
[2:52] * yanzheng (~zhyan@134.134.137.71) has joined #ceph
[2:54] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) has joined #ceph
[2:56] * grepory (~Adium@2600:1010:b016:9fb4:bce3:6388:3bff:1ce4) Quit (Quit: Leaving.)
[2:57] <sherry> size_t operator()(const sobject_t &r) const is used by which file?!
[2:59] * lordinvader (~lordinvad@14.139.82.6) has joined #ceph
[3:00] * yy-nm (~Thunderbi@220.184.128.218) has joined #ceph
[3:00] <joshd> it might not be used anymore - OSDMap::object_locator_to_pg() would be where I would expect to see it, but ceph_str_hash() is called directly there
[3:00] <lordinvader> hi, I'm facing some troubles trying to set up a very simple ceph configuration to manage block devices with libvirt - can I have both the admin node and server node on the same host?
[3:02] * rturk is now known as rturk-away
[3:03] * clayb (~kvirc@proxy-ny2.bloomberg.com) Quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
[3:06] <lordinvader> it needs to be a very minimal config for evaluation purposes only, and I'd prefer to set it up on one host only (my laptop)
[3:07] <dmick> lordinvader: yes
[3:11] <lordinvader> dmick, when I setup a monitor on my localhost, I don't get any keys in /var/lib/ceph/bootstrap-{mds-osd}
[3:11] * freedomhui (~freedomhu@117.79.232.248) has joined #ceph
[3:12] <lordinvader> if I run " sudo /usr/bin/python /usr/sbin/ceph-create-keys --cluster=ceph -i artoo " manually, I get an error - "INFO:ceph-create-keys:ceph-mon is not in quorum: u'probing'" (infinitely)
[3:14] * isaac_ (~isaac@mike-alien.esc.auckland.ac.nz) has joined #ceph
[3:18] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[3:18] * sagelap1 (~sage@2600:1012:b02d:5fe7:a1d8:6962:87a3:483b) Quit (Ping timeout: 480 seconds)
[3:18] <dmick> lordinvader: pastebin your ceph.conf?
[3:19] <lordinvader> dmick, http://paste.ubuntu.com/6064759/
[3:20] <dmick> mon initial members = 10 is not right
[3:20] <lordinvader> that's what ceph-deploy generated, should I change it to 1?
[3:20] <dmick> it's not a count
[3:21] <dmick> it's supposed to be a hostname. Perhaps your /etc/hosts doesn't map your hostname to a real IP?
[3:22] <nerdtron> lordinvader, how did you created/deployed your mons?
[3:22] <lordinvader> I'm not using a hostname - I'm using "10.1.33.13" or "ceph@10.1.33.13" (my IP assigned my local network)
[3:22] <nerdtron> it should look like this mon_initial_members = ceph-node1, ceph-node2, ceph-node3
[3:22] * ScOut3R (~ScOut3R@catv-89-133-25-52.catv.broadband.hu) has joined #ceph
[3:22] <lordinvader> I'm using ceph-deploy
[3:22] <lordinvader> It's unable to resolve my hostname (artoo) or localhost
[3:23] <nerdtron> each node should be able to resolve each other using hostnames
[3:23] <lordinvader> $ ceph-deploy mon create ceph@10.1.33.13
[3:23] <nerdtron> edit the /etc/hosts on each node and add the correct ip to hostname mappings of your nodes
[3:23] <nerdtron> wrong
[3:23] <nerdtron> sudo ceph-deploy mon create ceph-node1 ceph-node2 ceph-node3
[3:24] <nerdtron> it should be hostnames
[3:24] <dmick> (or in your case just the one)
[3:24] <nerdtron> did you follow the preflight check list?
[3:24] <dmick> nerdtron: sadly we don't talk about this there; I've filed a bug
[3:25] <nerdtron> dmick, about what? something is wrong on the preflight check list?
[3:25] <dmick> it's missing info about how hostnames and IPs need to work
[3:25] * haomaiwang (~haomaiwan@119.4.173.125) has joined #ceph
[3:25] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[3:25] <lordinvader> I couldn't understand some of the steps, but yes
[3:25] <lordinvader> I'll add a hostname and try it out - thanks!
[3:25] <dmick> lordinvader: the argument to ceph-deploy new and ceph-deploy mon create is "hostname"
[3:25] <nerdtron> oh yes,,, because when i followed it, it said "you should be able to ping the hostnames"
[3:25] <dmick> not user@host, not user@ip, etc.
[3:25] <nerdtron> like ping ceph-node1
[3:26] <nerdtron> i already assumed that /etc/hosts is already configured...
[3:26] <dmick> later ceph-deploy versions are getting better about validating that argument
[3:26] <nerdtron> lordinvader, you should be able to ping the hostname
[3:26] <dmick> nerdtron: the default installs often leave something less useful there. ping isn't enough
[3:26] <dmick> particularly to the localhost
[3:26] <dmick> which is this case
[3:27] <dmick> lordinvader: how did you get an installation where there is no local hostname?
[3:28] <lordinvader> there is a local hostname (artoo) - but ceph doesn't seem to be able to resolve it
[3:28] <dmick> what does getent hosts artoo say?
[3:28] <lordinvader> although it works just fine if I ping it
[3:28] <lordinvader> 127.0.1.1 artoo
[3:28] <nerdtron> lordinvader, ip of artoo?
[3:29] <lordinvader> eth0: 10.1.33.13, lo: 127.0.0.1
[3:29] <dmick> right. that's a "localhost"-style name (on 127/8), and that's not good enough
[3:29] <dmick> add another line to hosts: 10.1.33.13 artoo
[3:29] <dmick> then use artoo in the ceph-deploy commands
[3:29] * Qu310 (~Qten@ip-121-0-1-110.static.dsl.onqcomms.net) Quit (Remote host closed the connection)
[3:29] <lordinvader> ok, will try that out
[3:29] <nerdtron> yup add 10.1.33.13 artoo to /etc/hosts
[3:30] <nerdtron> then ceph-deploy mon install artoo
[3:30] <nerdtron> oopppss sorry
[3:30] <nerdtron> wrong type
[3:30] <lordinvader> should I remove the existing line '127.0.1.1 artoo' ?
[3:30] <dmick> no
[3:30] <dmick> it's fine to have two IPs
[3:31] * AfC (~andrew@2407:7800:200:1011:6e88:14ff:fe33:2a9c) Quit (Quit: Leaving.)
[3:31] <dmick> ceph requires a non-local one because in the more-common case of "multiple hosts" you don't want to use the local address
[3:31] * Qu310 (~Qten@ip-121-0-1-110.static.dsl.onqcomms.net) has joined #ceph
[3:31] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[3:32] <lordinvader> oh okay, let me retry the installation process. Thanks a ton :)
[3:33] * AfC (~andrew@2407:7800:200:1011:6e88:14ff:fe33:2a9c) has joined #ceph
[3:33] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[3:36] <lordinvader> yes, got the keyring files!! Looks like I might get the assignment done after all! :D :D
[3:36] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[3:36] * dmick goes to make his bug urgent
[3:37] <nerdtron> well i say it's not a bug but more like, assume you have your host files configured
[3:37] <dmick> nerdtron: it's a special configuration only needed by ceph
[3:38] <dmick> and not configured by the OS install
[3:38] <dmick> and it also need not necessarily be the host file; there are other sources of name information
[3:39] <dmick> gah. maybe I never actually filed it
[3:39] <nerdtron> yes, but it already said in the tutorial, on the lower part of the preflight check list, there is a section on how to configure the hostname resolution
[3:40] <dmick> where?
[3:41] <nerdtron> http://ceph.com/docs/master/start/quick-start-preflight/#hostname-resolution
[3:41] <dmick> oh, in the "Getting Started", but not in the path at http://ceph.com/docs/master/rados/deployment/
[3:41] <dmick> how annoying
[3:41] * bandrus (~Adium@cpe-76-95-217-129.socal.res.rr.com) Quit (Quit: Leaving.)
[3:42] <nerdtron> yeah..i'm lucky it's the first thing i read
[3:42] <dmick> they're both called "preflight checklist" too. awesome.
[3:42] <dmick> well yeah, something like that, although that could certainly be worded better too
[3:42] <dmick> and doesn't have to be ipv4
[3:43] * LeaChim (~LeaChim@97e00ac2.skybroadband.com) Quit (Ping timeout: 480 seconds)
[3:44] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[3:46] * markbby (~Adium@168.94.245.2) has joined #ceph
[3:47] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[3:48] <dmick> http://tracker.ceph.com/issues/6234
[3:49] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[3:52] * sherry (~sherry@wireless-nat-10.auckland.ac.nz) Quit (Quit: Konversation terminated!)
[3:52] <nerdtron> dmick, good move
[3:53] <lordinvader> :)
[4:00] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[4:03] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[4:07] <alphe> ok ceph cluster finished to full reinstall :)
[4:08] <alphe> but I think I could use ceph-deploy activate on my nodes to solve my osd disapearation problems
[4:08] * rovar (~oftc-webi@69.191.241.59) Quit (Remote host closed the connection)
[4:08] * markbby (~Adium@168.94.245.2) Quit (Quit: Leaving.)
[4:12] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) Quit (Quit: Leaving.)
[4:13] <alphe> forgot to set up the mds
[4:13] <alphe> and now the mds doesn t want
[4:15] <alphe> ceph-deploy 1.2.3 fails to deploy mds !
[4:15] <alphe> ceph-deploy 1.2.3 fails to create mds !
[4:16] <alphe> '/var/lib/ceph/bootstrap-mds/ceph.keyring.26951.tmp'
[4:16] <alphe> no such file or directory
[4:16] <alphe> and it is not existing in fact ...
[4:18] <alphe> problem with /var/lib/ceph/mds/ceph-mds01
[4:18] <alphe> too
[4:18] <alphe> so manually mkdir -p /var/lib/ceph/mds/ceph-mds01
[4:18] <alphe> mkdir -p /var/lib/ceph/bootstrap-mds
[4:18] <alphe> then call again ceph-deploy mds create hostname and bang that works
[4:19] <alphe> mdsmap e4: 1/1/1 up {0=mds01=up:active}
[4:19] * julian (~julianwa@125.70.133.187) has joined #ceph
[4:20] <alphe> ok folks lets call it a day !
[4:20] <alphe> goodbye see you soon enough :)
[4:21] * alphe (~alphe@0001ac6f.user.oftc.net) Quit (Quit: Leaving)
[4:25] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[4:25] * aliguori (~anthony@cpe-70-112-153-179.austin.res.rr.com) Quit (Remote host closed the connection)
[4:26] * sjm_ (~sjm@38.122.20.226) has joined #ceph
[4:32] * sjm (~sjm@2607:f298:a:607:5165:a808:d1c7:3e6f) Quit (Ping timeout: 480 seconds)
[4:33] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[4:42] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) has joined #ceph
[4:45] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) Quit (Quit: sprachgenerator)
[4:45] * madkiss (~madkiss@2001:6f8:12c3:f00f:5c8c:d54d:43b5:1a95) Quit (Ping timeout: 480 seconds)
[4:45] * julian (~julianwa@125.70.133.187) Quit (Quit: afk)
[4:45] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[4:48] * xarses (~andreww@c-71-202-167-197.hsd1.ca.comcast.net) has joined #ceph
[4:50] * AfC (~andrew@2407:7800:200:1011:6e88:14ff:fe33:2a9c) Quit (Quit: Leaving.)
[4:54] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) Quit (Ping timeout: 480 seconds)
[4:56] * Cube (~Cube@12.248.40.138) Quit (Quit: Leaving.)
[5:01] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[5:06] * fireD (~fireD@93-139-138-90.adsl.net.t-com.hr) has joined #ceph
[5:06] * jlogan1 (~Thunderbi@2600:c00:3010:1:1::40) Quit (Ping timeout: 480 seconds)
[5:07] * fireD_ (~fireD@93-139-182-124.adsl.net.t-com.hr) Quit (Ping timeout: 480 seconds)
[5:12] * sjm_ (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[5:17] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[5:25] * lordinvader (~lordinvad@14.139.82.6) Quit (Ping timeout: 480 seconds)
[5:25] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[5:29] * jlogan1 (~Thunderbi@2600:c00:3010:1:1::40) has joined #ceph
[5:29] <nigwil> Jao asked about the unexpected ceph-deploy problem I had a couple of days ago, I have dug through my shell history and capture what I think is the sequence of events and the configuration characteristics that might have contributed and put them here: http://pastebin.com/LZ7wGCYe
[5:34] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[5:36] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[5:40] * gregaf1 (~Adium@cpe-172-250-66-79.socal.res.rr.com) Quit (Quit: Leaving.)
[5:40] * gregaf1 (~Adium@cpe-172-250-66-79.socal.res.rr.com) has joined #ceph
[5:40] * gregaf1 (~Adium@cpe-172-250-66-79.socal.res.rr.com) Quit ()
[5:44] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[6:10] <malcolm> Sorry to be a pest, but I've got a couple questions about performance relitive to native disk.
[6:12] * wschulze (~wschulze@cpe-69-203-80-81.nyc.res.rr.com) has joined #ceph
[6:13] * sherry (~sherry@wireless-nat-10.auckland.ac.nz) has joined #ceph
[6:14] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[6:14] * haomaiwang (~haomaiwan@119.4.173.125) Quit (Read error: Connection reset by peer)
[6:15] * haomaiwang (~haomaiwan@119.4.173.125) has joined #ceph
[6:16] <nigwil> hi malcolm, think most of the Ceph experts are now asleep (late in north america, and europe is not yet awake)
[6:17] <sherry> what is the command to get the version of ceph developer after is was cloned?
[6:18] * sagelap (~sage@76.89.177.113) has joined #ceph
[6:19] <sherry> what is the command to get the version of ceph developer after is was cloned?
[6:20] * grepory (~Adium@239.sub-70-197-5.myvzw.com) has joined #ceph
[6:20] <nigwil> sherry: sorry I don't understand the question, you want the version of ceph running? (ceph --version)
[6:21] <sherry> not that, the version of the source codess
[6:22] <sherry> because I got the newest version .68 but in some files like acconfig.h, it shows the version is still .67!
[6:22] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[6:26] <sage> sherry: where do you see that?
[6:26] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[6:26] <sage> sherry: acconfig.h is only rebuild whenyou run make..
[6:27] <sage> grep AC_INIT configure.ac
[6:28] <nigwil> malcolm: Sage is here, you should ask your question! :-)
[6:30] * grepory (~Adium@239.sub-70-197-5.myvzw.com) Quit (Quit: Leaving.)
[6:31] <sherry> /* Version number of package */
[6:31] <sherry> #define VERSION "0.67"
[6:32] <sherry> it should make the new one, right?
[6:33] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) Quit (Ping timeout: 480 seconds)
[6:34] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[6:34] <sage> sherry: git describe
[6:35] <sage> sherry: grep AC_INIT configure.ac
[6:37] * sjusthm (~sam@24.205.35.233) Quit (Ping timeout: 480 seconds)
[6:47] * portante is now known as portante|afk
[6:49] <sherry> I assumed running "git submodule update --init" should update the codes, which didnt!
[6:56] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[6:57] * ScOut3R (~ScOut3R@catv-89-133-25-52.catv.broadband.hu) Quit (Read error: Operation timed out)
[7:04] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[7:09] * sherry (~sherry@wireless-nat-10.auckland.ac.nz) Quit (Read error: Connection reset by peer)
[7:09] * sherry (~sherry@wireless-nat-10.auckland.ac.nz) has joined #ceph
[7:19] * lx0 (~aoliva@lxo.user.oftc.net) has joined #ceph
[7:19] * lxo (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[7:24] <yy-nm> sage: how way in to get know rbd working better?
[7:25] * shang (~ShangWu@175.41.48.77) Quit (Ping timeout: 480 seconds)
[7:33] * isaac_ (~isaac@mike-alien.esc.auckland.ac.nz) Quit (Ping timeout: 480 seconds)
[7:39] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[7:46] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Ping timeout: 480 seconds)
[7:50] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[7:52] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[7:54] * sachanab (~oftc-webi@zswa01cs004-da01.atlanta.hp.com) has joined #ceph
[7:55] <sachanab> anybody there??
[7:55] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) has joined #ceph
[7:56] <nerdtron> yo!
[7:56] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[7:57] <sachanab> i am trying to setup a cluster using cuttlefish version
[7:57] <sachanab> and when i issue "ceph-deploy mon create <hostname>
[7:57] <sachanab> i get following errors
[7:57] <sachanab> tea][ERROR ] Traceback (most recent call last): [tea][ERROR ] File "/usr/lib/python2.6/site-packages/ceph_deploy/hosts/centos/mon/create.py", line 18, in create [tea][ERROR ] File "/usr/lib/python2.6/site-packages/ceph_deploy/util/decorators.py", line 10, in inner [tea][ERROR ] def inner(*args, **kwargs): [tea][ERROR ] File "/usr/lib/python2.6/site-packages/ceph_deploy/util/wrappers.py", line 6, in remote_call [tea][ERROR ]
[7:58] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Read error: No route to host)
[7:58] <sachanab> @nerdtron
[7:58] <cephalobot> sachanab: Error: "nerdtron" is not a valid command.
[7:59] <sachanab> tron, any ideas?
[8:02] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[8:04] * sleinen1 (~Adium@2001:620:0:26:1859:c019:f7a1:fdc2) has joined #ceph
[8:07] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[8:07] * ChanServ sets mode +v andreask
[8:07] * sleinen1 (~Adium@2001:620:0:26:1859:c019:f7a1:fdc2) Quit ()
[8:08] * sleinen1 (~Adium@77-58-245-10.dclient.hispeed.ch) has joined #ceph
[8:10] * sleinen (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[8:11] <malcolm> Sorry for the delay in asking, I hope Sage is still around... anyway here it goes...just a quick one no rush on the answer/comments. I was testing CephFS and had some striping values set. I expected to see performance greater than one disk, was that an incorrect expctation? my stripe unit was quite long and my objects quite large. Count was 12 (24 disks in the cluster)
[8:13] * yy-nm (~Thunderbi@220.184.128.218) Quit (Quit: yy-nm)
[8:16] * sleinen1 (~Adium@77-58-245-10.dclient.hispeed.ch) Quit (Ping timeout: 480 seconds)
[8:16] * wschulze (~wschulze@cpe-69-203-80-81.nyc.res.rr.com) Quit (Quit: Leaving.)
[8:17] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[8:19] * yy-nm (~Thunderbi@220.184.128.218) has joined #ceph
[8:29] * sleinen (~Adium@2001:620:0:25:e081:2938:5483:bdc3) has joined #ceph
[8:33] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[8:33] * ChanServ sets mode +v andreask
[8:36] * Vjarjadian (~IceChat77@176.254.37.210) Quit (Quit: Relax, its only ONES and ZEROS!)
[8:37] * NaioN (stefan@andor.naion.nl) Quit (Quit: Lost terminal)
[8:39] * NaioN (stefan@andor.naion.nl) has joined #ceph
[8:43] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[8:49] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Read error: Operation timed out)
[8:50] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[8:59] * JustEra (~JustEra@ALille-555-1-86-217.w90-7.abo.wanadoo.fr) Quit (Quit: This computer has gone to sleep)
[9:00] * sherry (~sherry@wireless-nat-10.auckland.ac.nz) Quit (Quit: Konversation terminated!)
[9:01] * odyssey4me (~odyssey4m@165.233.71.2) has joined #ceph
[9:05] * thelan (~thelan@paris.servme.fr) Quit (Read error: Operation timed out)
[9:14] * grepory (~Adium@12.236.17.3) has joined #ceph
[9:16] * Bada (~Bada@dom95-4-88-185-75-36.fbx.proxad.net) has joined #ceph
[9:16] * JustEra (~JustEra@89.234.148.11) has joined #ceph
[9:17] * grepory1 (~Adium@2600:1010:b003:6e9c:6dc3:e0b8:e2a0:ceb2) has joined #ceph
[9:17] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Read error: Operation timed out)
[9:20] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[9:20] * BManojlovic (~steki@91.195.39.5) has joined #ceph
[9:21] <nigwil> it is after midnight in Inktank-land so I expect they're are off to the land of nod :-)
[9:23] * grepory (~Adium@12.236.17.3) Quit (Ping timeout: 480 seconds)
[9:25] * malcolm (~malcolm@silico24.lnk.telstra.net) Quit (Ping timeout: 480 seconds)
[9:25] * freedomhui (~freedomhu@117.79.232.248) Quit (Quit: Leaving...)
[9:28] * thelan (~thelan@paris.servme.fr) has joined #ceph
[9:32] * Bada (~Bada@dom95-4-88-185-75-36.fbx.proxad.net) Quit (Ping timeout: 480 seconds)
[9:34] * pieter_ (~pieter@105-236-213-182.access.mtnbusiness.co.za) has joined #ceph
[9:35] * freedomhui (~freedomhu@li565-182.members.linode.com) has joined #ceph
[9:35] <pieter_> Hi guys. Does RAID1 make sense for a ceph OSD device, or might 2x ceph OSD's instead do the same job in terms of redundancy?
[9:36] <pieter_> (so justr 2x ceph osds in total each being a raid1, or 4x 2TB OSD's without raid?)
[9:36] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[9:36] * ChanServ sets mode +v andreask
[9:37] * shang (~ShangWu@59-124-245-140.HINET-IP.hinet.net) has joined #ceph
[9:37] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[9:37] * freedomhu (~freedomhu@li565-182.members.linode.com) has joined #ceph
[9:37] * freedomhui (~freedomhu@li565-182.members.linode.com) Quit (Read error: No route to host)
[9:38] <nigwil> my understanding is that RAID never makes sense for OSDs, you're relying on Ceph to make replicas of objects anyway, and having raw-device OSDs maximises IO performance (RAID is going to add latency)
[9:43] * freedomhu (~freedomhu@li565-182.members.linode.com) Quit (Quit: Leaving...)
[9:45] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[9:45] * freedomhui (~freedomhu@211.155.113.212) has joined #ceph
[9:46] <nigwil> pieter_: there are responses on the mailing list about OSDs and RAID: http://search.gmane.org/?query=raid+osd&author=&group=gmane.comp.file-systems.ceph.user&sort=relevance&DEFAULTOP=and&xP=Zraid&xFILTERS=Gcomp.file-systems.ceph.user---A
[9:46] * jlogan1 (~Thunderbi@2600:c00:3010:1:1::40) Quit (Ping timeout: 480 seconds)
[9:46] * LeaChim (~LeaChim@97e00ac2.skybroadband.com) has joined #ceph
[9:48] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Read error: Operation timed out)
[9:50] <pieter_> Thanks nigwil
[9:50] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[9:55] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Quit: Leaving.)
[9:56] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[9:56] * ChanServ sets mode +v andreask
[9:57] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit ()
[9:57] * grepory1 (~Adium@2600:1010:b003:6e9c:6dc3:e0b8:e2a0:ceb2) Quit (Quit: Leaving.)
[9:57] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[9:57] * ChanServ sets mode +v andreask
[9:57] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit ()
[9:58] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[9:58] * ChanServ sets mode +v andreask
[10:00] * grepory (~Adium@2600:1010:b02b:126c:258d:bf80:fbdb:3b3e) has joined #ceph
[10:08] * mattt (~mattt@92.52.76.140) has joined #ceph
[10:08] * jbd_ (~jbd_@2001:41d0:52:a00::77) has joined #ceph
[10:08] * grepory1 (~Adium@12.236.17.3) has joined #ceph
[10:09] * sachanab (~oftc-webi@zswa01cs004-da01.atlanta.hp.com) Quit (Quit: Page closed)
[10:10] * thomnico (~thomnico@2a01:e35:8b41:120:dc85:9784:3cef:7288) Quit (Ping timeout: 480 seconds)
[10:10] * PITon (~pavel@195.182.195.107) has joined #ceph
[10:14] * grepory (~Adium@2600:1010:b02b:126c:258d:bf80:fbdb:3b3e) Quit (Ping timeout: 480 seconds)
[10:16] * thomnico (~thomnico@2a01:e35:8b41:120:74c0:e849:3e:503b) has joined #ceph
[10:24] * ShaunR- (~ShaunR@staff.ndchost.com) Quit (Read error: Connection reset by peer)
[10:24] * ShaunR (~ShaunR@staff.ndchost.com) has joined #ceph
[10:24] * allsystemsarego (~allsystem@188.25.134.187) has joined #ceph
[10:25] * capri_on (~capri@212.218.127.222) has joined #ceph
[10:27] * ivan` (~ivan`@000130ca.user.oftc.net) Quit (Ping timeout: 480 seconds)
[10:31] * ivan` (~ivan`@000130ca.user.oftc.net) has joined #ceph
[10:32] * capri (~capri@212.218.127.222) Quit (Ping timeout: 480 seconds)
[10:32] * Qu310 (~Qten@ip-121-0-1-110.static.dsl.onqcomms.net) Quit (Remote host closed the connection)
[10:40] <jerker> nigwil: hmmm RAID do not necessarily add latency. it does for RAID5/6 for smaller writes than a full write and if the battery cache cant cope with all writes and the drives are not fast enough... http://en.wikipedia.org/wiki/Standard_RAID_levels#Latency
[10:41] <jerker> nigwil: for RAID1 it should not add latency, as far as i understand
[10:41] * Qu310 (~Qten@ip-121-0-1-110.static.dsl.onqcomms.net) has joined #ceph
[10:43] <jerker> nigwil: but in general, the redundancy might be worse with combining local redundancy (RAID1/5/6/Z1/Z2/Z3etc) and replication than just using all spare drives into Ceph. I have not really calculated this. One can, I guess.
[10:43] * grepory1 (~Adium@12.236.17.3) Quit (Quit: Leaving.)
[10:43] <jerker> nigwil: But if Ceph get Erasure coding it will be better to keep everything in Ceph for sure.
[10:45] <nigwil> fair point. I would wonder how effective the sync of angular rotation is in practice (something for me to research sometime).
[10:46] <nigwil> Another aspect to consider is recovery/rebuild times. A failed OSD means that just its actual content needs to be re-replicated (rather than the total capacity of the device as per RAID)
[10:46] * AfC (~andrew@2407:7800:200:1011:2ad2:44ff:fe08:a4c) Quit (Ping timeout: 480 seconds)
[10:49] * JustEra (~JustEra@89.234.148.11) Quit (Ping timeout: 480 seconds)
[10:49] <jerker> nigwil: yes. i think netapp used to keep track on that. some old sun systems also. I dont know about nowadays.
[10:50] <nigwil> I've seen rotational sync been implemented on SMD drives in 1987, so it is an old idea :-)
[10:50] * JustEra (~JustEra@89.234.148.11) has joined #ceph
[11:01] * yy-nm (~Thunderbi@220.184.128.218) Quit (Quit: yy-nm)
[11:02] <jerker> nowadays: just add more SSDs :)
[11:03] * mschiff (~mschiff@p4FD7F716.dip0.t-ipconnect.de) has joined #ceph
[11:03] * yanzheng (~zhyan@134.134.137.71) Quit (Remote host closed the connection)
[11:06] * wogri_risc (~wogri_ris@2a00:dc0:f:f001:f073:c699:7503:827a) has joined #ceph
[11:10] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Read error: Connection reset by peer)
[11:11] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[11:27] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) Quit (Quit: tryggvil)
[11:32] * thomnico (~thomnico@2a01:e35:8b41:120:74c0:e849:3e:503b) Quit (Ping timeout: 480 seconds)
[11:36] * DarkAce-Z (~BillyMays@50.107.55.36) has joined #ceph
[11:38] * DarkAceZ (~BillyMays@50.107.55.36) Quit (Ping timeout: 480 seconds)
[11:40] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) has joined #ceph
[11:46] * tryggvil (~tryggvil@178.19.53.254) has joined #ceph
[11:53] * smiley_ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) Quit (Quit: smiley_)
[12:00] * tziOm (~bjornar@194.19.106.242) has joined #ceph
[12:00] * freedomhui (~freedomhu@211.155.113.212) Quit (Quit: Leaving...)
[12:01] * freedomhui (~freedomhu@117.79.232.248) has joined #ceph
[12:04] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Ping timeout: 480 seconds)
[12:05] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) has joined #ceph
[12:12] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[12:12] * ChanServ sets mode +v andreask
[12:16] * ScOut3R (~ScOut3R@catv-89-133-17-71.catv.broadband.hu) has joined #ceph
[12:17] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Read error: Operation timed out)
[12:20] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[12:21] * nerdtron (~kenneth@202.60.8.252) Quit (Remote host closed the connection)
[12:45] * smiley_ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) has joined #ceph
[12:48] * wogri_risc (~wogri_ris@2a00:dc0:f:f001:f073:c699:7503:827a) Quit (Quit: wogri_risc)
[12:56] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) has joined #ceph
[12:57] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) Quit (Ping timeout: 480 seconds)
[12:58] * Midnightmyth_ (~quassel@93-167-84-102-static.dk.customer.tdc.net) has joined #ceph
[13:03] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit (Read error: Operation timed out)
[13:05] * Midnightmyth (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[13:05] * wogri_risc (~wogri_ris@85.233.126.171) has joined #ceph
[13:05] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) has joined #ceph
[13:06] * john_barbee_ (~jbarbee@c-98-220-74-174.hsd1.in.comcast.net) Quit ()
[13:07] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Quit: Leaving.)
[13:07] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[13:07] * ChanServ sets mode +v andreask
[13:08] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit ()
[13:10] * haomaiwang (~haomaiwan@119.4.173.125) Quit (Remote host closed the connection)
[13:10] * haomaiwang (~haomaiwan@li498-162.members.linode.com) has joined #ceph
[13:14] * haomaiwa_ (~haomaiwan@119.4.173.125) has joined #ceph
[13:18] * mikedawson (~chatzilla@c-98-220-189-67.hsd1.in.comcast.net) Quit (Quit: ChatZilla 0.9.90.1 [Firefox 23.0.1/20130814063812])
[13:18] * allsystemsarego (~allsystem@188.25.134.187) Quit (Quit: Leaving)
[13:21] * haomaiwang (~haomaiwan@li498-162.members.linode.com) Quit (Ping timeout: 480 seconds)
[13:27] * rendar (~s@host53-178-dynamic.36-79-r.retail.telecomitalia.it) has joined #ceph
[13:34] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) has joined #ceph
[13:34] * ChanServ sets mode +v andreask
[13:38] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[13:45] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) has joined #ceph
[13:46] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[13:47] * rendar (~s@host53-178-dynamic.36-79-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[13:48] * allsystemsarego (~allsystem@188.25.134.187) has joined #ceph
[13:49] * alfredodeza (~alfredode@c-24-131-46-23.hsd1.ga.comcast.net) has joined #ceph
[13:49] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) has joined #ceph
[13:50] * kraken (~kraken@c-24-131-46-23.hsd1.ga.comcast.net) has joined #ceph
[13:50] * shang (~ShangWu@59-124-245-140.HINET-IP.hinet.net) Quit (Read error: Operation timed out)
[13:54] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) has joined #ceph
[13:54] * jeff-YF (~jeffyf@pool-173-66-21-43.washdc.fios.verizon.net) has joined #ceph
[13:57] * yanzheng (~zhyan@134.134.139.70) has joined #ceph
[14:00] * shang (~ShangWu@111.80.58.237) has joined #ceph
[14:04] * tserong__ is now known as tserong
[14:11] * jeff-YF_ (~jeffyf@67.23.123.228) has joined #ceph
[14:16] * jeff-YF (~jeffyf@pool-173-66-21-43.washdc.fios.verizon.net) Quit (Ping timeout: 480 seconds)
[14:16] * jeff-YF_ is now known as jeff-YF
[14:20] * markbby (~Adium@168.94.245.1) has joined #ceph
[14:29] * shang (~ShangWu@111.80.58.237) Quit (Read error: Connection reset by peer)
[14:32] * KevinPerks (~Adium@cpe-066-026-239-136.triad.res.rr.com) Quit (Ping timeout: 480 seconds)
[14:34] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) Quit (Ping timeout: 480 seconds)
[14:34] * libsysguy (~libsysguy@2620:0:28a0:2004:b07b:b238:3dda:e31b) has joined #ceph
[14:35] * nhorman (~nhorman@hmsreliant.think-freely.org) has joined #ceph
[14:36] <libsysguy> so I am trying to come up with the best way to distribute ceph over a WAN
[14:36] <libsysguy> would it be best to have multiple clusters that sync with an external script?
[14:36] <libsysguy> it seems like nodes over a wan can be fickle
[14:38] * jeff-YF (~jeffyf@67.23.123.228) Quit (Quit: jeff-YF)
[14:47] * yanzheng (~zhyan@134.134.139.70) Quit (Ping timeout: 480 seconds)
[14:52] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[14:54] * haomaiwa_ (~haomaiwan@119.4.173.125) Quit (Remote host closed the connection)
[14:54] * portante|afk is now known as portante
[14:56] * haomaiwang (~haomaiwan@119.6.74.59) has joined #ceph
[14:57] * smiley_ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) Quit (Quit: smiley_)
[14:57] <wogri_risc> libsysguy
[14:57] <wogri_risc> I was in the same situation once :)
[14:57] <wogri_risc> how "slow" and "delayish" is your wan?
[14:57] <libsysguy> and you made it out alive wogri_risc
[14:57] <wogri_risc> you didn't see me bleeding :)
[14:58] <libsysguy> well it depends
[14:58] <wogri_risc> answer is: it really depends on your use case.
[14:58] <libsysguy> heh jinx
[14:58] <wogri_risc> if you want to use radosgw - there is something implemented in dumpling that supports multiple ceph clusters for intra-cluster-replication
[14:58] <libsysguy> well that is promising and unknown
[14:59] <wogri_risc> if you want to use rbd or rados or cephfs you're better off using separate clusters
[14:59] <libsysguy> I am currently testing for use with the radosgw
[15:00] <libsysguy> is the intra-cluster-replication the bit about messing with the crush map and the heartbeat?
[15:00] <wogri_risc> no
[15:00] * libsysguy breaths a sigh of relief
[15:00] <wogri_risc> sorry, i meant inter-cluster replication (you replicate between two clusters)
[15:01] <libsysguy> gotcha
[15:01] <libsysguy> do you know where that is in the docs?
[15:01] <libsysguy> google is my friend, but not this day
[15:02] <wogri_risc> yeah, have troubles finding it myself.
[15:02] <wogri_risc> I hope I didn't promise too much (don't use rgw myself)
[15:02] <libsysguy> I see some vague mention of it in the release notes
[15:03] <libsysguy> its probably just under-documented at this point if its a new feature in dumpling
[15:03] * psiekl (psiekl@wombat.eu.org) Quit (Read error: Connection reset by peer)
[15:03] * psiekl (psiekl@wombat.eu.org) has joined #ceph
[15:03] <wogri_risc> maybe here: http://ceph.com/docs/master/radosgw/config-ref/#regions
[15:03] <wogri_risc> I'm a little short on time and have to move on
[15:04] <wogri_risc> but the best place to ask this is the ceph-users Mailinglist
[15:04] <wogri_risc> you will be helped
[15:04] <libsysguy> sure, thanks for your time wogri_risc++
[15:05] <LCF> wogri_risc: with regions is it possible to run DR DC with copy of all files ?
[15:06] * wogri_risc (~wogri_ris@85.233.126.171) has left #ceph
[15:12] * KevinPerks (~Adium@cpe-066-026-252-218.triad.res.rr.com) has joined #ceph
[15:15] * KevinPerks1 (~Adium@cpe-066-026-252-218.triad.res.rr.com) has joined #ceph
[15:15] * KevinPerks (~Adium@cpe-066-026-252-218.triad.res.rr.com) Quit (Read error: Connection reset by peer)
[15:15] * clayb (~kvirc@proxy-nj1.bloomberg.com) has joined #ceph
[15:18] * KevinPerks1 (~Adium@cpe-066-026-252-218.triad.res.rr.com) has left #ceph
[15:18] * wschulze (~wschulze@cpe-69-203-80-81.nyc.res.rr.com) has joined #ceph
[15:19] * markbby (~Adium@168.94.245.1) Quit (Quit: Leaving.)
[15:19] * markbby (~Adium@168.94.245.1) has joined #ceph
[15:26] * grepory (~Adium@12.236.17.3) has joined #ceph
[15:30] * smiley_ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) has joined #ceph
[15:34] * ScOut3R (~ScOut3R@catv-89-133-17-71.catv.broadband.hu) Quit (Remote host closed the connection)
[15:34] * ScOut3R (~ScOut3R@catv-89-133-17-71.catv.broadband.hu) has joined #ceph
[15:37] * aliguori (~anthony@cpe-70-112-153-179.austin.res.rr.com) has joined #ceph
[15:38] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[15:43] * jskinner (~jskinner@199.127.136.233) has joined #ceph
[15:46] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[15:47] * dmsimard (~Adium@108.163.152.2) has joined #ceph
[15:48] * jeff-YF (~jeffyf@67.23.117.122) has joined #ceph
[15:48] * KevinPerks1 (~Adium@cpe-066-026-252-218.triad.res.rr.com) has joined #ceph
[15:53] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[15:58] * sjusthm (~sam@24-205-35-233.dhcp.gldl.ca.charter.com) has joined #ceph
[16:03] * tziOm (~bjornar@194.19.106.242) Quit (Remote host closed the connection)
[16:11] * PerlStalker (~PerlStalk@2620:d3:8000:192::70) has joined #ceph
[16:11] * andreask (~andreask@h081217135028.dyn.cm.kabsi.at) Quit (Quit: Leaving.)
[16:13] * MACscr (~Adium@c-98-214-103-147.hsd1.il.comcast.net) has joined #ceph
[16:14] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) has joined #ceph
[16:21] * ScOut3R_ (~ScOut3R@catv-89-133-17-71.catv.broadband.hu) has joined #ceph
[16:26] * rendar (~s@host176-181-dynamic.10-87-r.retail.telecomitalia.it) has joined #ceph
[16:26] * ScOut3R (~ScOut3R@catv-89-133-17-71.catv.broadband.hu) Quit (Ping timeout: 480 seconds)
[16:28] * themgt_ (~themgt@201-223-223-205.baf.movistar.cl) has joined #ceph
[16:31] * themgt (~themgt@201-223-248-202.baf.movistar.cl) Quit (Ping timeout: 480 seconds)
[16:31] * themgt_ is now known as themgt
[16:36] <loicd> joao: while attending http://devlog.cnrs.fr/jdev2013/t5.a4 today I've been told that Paxos inventor http://en.wikipedia.org/wiki/Leslie_Lamport lives in Paris !
[16:36] * BillK (~BillK-OFT@58-7-131-166.dyn.iinet.net.au) Quit (Ping timeout: 480 seconds)
[16:38] <dmsimard> From what I understand, when you mount a block device with RBD, you have to use a monitor as an "endpoint". For instance, "rbd map <imagename> --pool <poolname> --name client.admin -m <monitor". My question is, does the client keep some sort of connection opened with the monitor or is it stateless ? Could I perhaps point to a load balancer in front of my monitors instead ?
[16:38] <loicd> I wrote and ask to meet to discuss about Ceph. An plausible excuse to get a chance to meet the legend. I only knew his name because of latex.
[16:39] <jmlowe> dmsimard: I usually put ceph.conf on my hosts then I don't need to specify the monitor
[16:39] <dmsimard> jmlowe: Fair enough, so what does your configuration point to ? A specific monitor ? What happens if that monitor fails ?
[16:40] <jmlowe> dmsimard: I have all my monitors in my conf, tries till it finds a working one
[16:41] <dmsimard> jmlowe: Oh, interesting - I wasn't aware of that behaviour. Thanks for that.
[16:41] * shang (~ShangWu@210.242.156.234) has joined #ceph
[16:41] <jmlowe> dmsimard: I have vague memories of using a comma separated list when giving a monitor argument, if that's a real memory then you should be able to stick all of your monitors in that same argument
[16:42] * vata (~vata@2607:fad8:4:6:ad7b:c365:d5eb:b7b2) has joined #ceph
[16:42] <jmlowe> dmsimard: And I believe that once connected to a monitor a client will learn about the other monitors
[16:42] <dmsimard> Makes sense.
[16:43] <alfredodeza> sprachgenerator: do you happen to have mon logs for the problem from yesterday?
[16:44] * mtanski (~mtanski@69.193.178.202) has joined #ceph
[16:44] * libsysguy (~libsysguy@2620:0:28a0:2004:b07b:b238:3dda:e31b) Quit (Quit: Leaving.)
[16:45] * libsysguy (~libsysguy@ng1.cptxoffice.net) has joined #ceph
[16:45] * libsysguy (~libsysguy@ng1.cptxoffice.net) Quit ()
[16:46] * sagelap1 (~sage@2600:1012:b024:4c7c:f836:7a76:e39:2411) has joined #ceph
[16:46] <sagelap1> nhm: the plana workers are probably still busy with the stuff from last tnight
[16:46] <sagelap1> watch 'teuthology-lock --list -a --brief --machine-type plana | sort -k +4'
[16:47] <sagelap1> mostly the upgrade suite
[16:49] <nhm> sagelap1: I guess the behavior confuses me, I thought if the nodes weren't free I'd still see a directory with a teuthology.log file saying it's waiting for nodes to free up.
[16:51] * sagelap (~sage@76.89.177.113) Quit (Ping timeout: 480 seconds)
[16:55] <loicd> speaking of teuthology ... I asked my company for 20 machines to run teuthology in the same way inktank does ( well, that's the ultimate goal anyway ;-). Management asked : why 20 ? This is a lot. I replied that inktank has hundreds but we don't need to run all the suites ( no interest in cephfs at the moment ) and we're only two developers. Therefore 20 should be enough for both teuthology + gitbuilder. Is it a realistic figure ?
[16:58] * markbby (~Adium@168.94.245.1) Quit (Remote host closed the connection)
[16:59] * markbby (~Adium@168.94.245.1) has joined #ceph
[17:03] * grepory (~Adium@12.236.17.3) Quit (Quit: Leaving.)
[17:05] * sprachgenerator (~sprachgen@c-50-141-192-36.hsd1.il.comcast.net) Quit (Quit: sprachgenerator)
[17:06] <sagelap1> nhm: mostly; there are a fixed number of workers running (~80?) and the queue can be bigger than that
[17:06] * JustEra (~JustEra@89.234.148.11) Quit (Ping timeout: 480 seconds)
[17:07] <sagelap1> loicd: yeah; you can also use our gitbuilders.
[17:07] <sagelap1> loicd: also, we'd like to make teuthology able to allocate machines using euca2ools (on openstack or aws or whatever) which would avoid dedicated machines..
[17:08] <sagelap1> loicd: honestly i think that is the easiest way to get more teuthology users. anybody can use aws, and lots of people already have openstack clusters for dev/test (and production)
[17:08] * grepory (~Adium@12.236.17.3) has joined #ceph
[17:09] * jlogan (~Thunderbi@2600:c00:3010:1:1::40) has joined #ceph
[17:09] * wogri_risc (~wogri_ris@85.233.124.80) has joined #ceph
[17:11] * Vjarjadian (~IceChat77@176.254.37.210) has joined #ceph
[17:11] * pieter_ (~pieter@105-236-213-182.access.mtnbusiness.co.za) Quit (Quit: Konversation terminated!)
[17:12] <nhm> sagelap1: is that a fixed number of workers per machine class? IE the burnupi/mira suite showed up immediately in /a, but plana runs inbetween them still haven't showed up.
[17:13] <loicd> sagelap1: understood. Do 20 virtual machines with 32GB RAM each + 4 core + 100GB disk seem overkill to run an independant teuthology ?
[17:15] * grepory (~Adium@12.236.17.3) Quit (Read error: Connection reset by peer)
[17:15] * grepory (~Adium@12.236.17.3) has joined #ceph
[17:17] * diegows (~diegows@200.68.116.185) has joined #ceph
[17:18] * zackc is excited to start working on teuthology-as-an-openstack-user
[17:19] * jeff-YF (~jeffyf@67.23.117.122) Quit (Quit: jeff-YF)
[17:25] * sagelap2 (~sage@38.122.20.226) has joined #ceph
[17:25] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) has joined #ceph
[17:25] * themgt (~themgt@201-223-223-205.baf.movistar.cl) Quit (Quit: Pogoapp - http://www.pogoapp.com)
[17:27] * haomaiwang (~haomaiwan@119.6.74.59) Quit (Remote host closed the connection)
[17:28] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) Quit (Ping timeout: 480 seconds)
[17:28] * jeff-YF (~jeffyf@216.14.83.26) has joined #ceph
[17:29] * sagelap1 (~sage@2600:1012:b024:4c7c:f836:7a76:e39:2411) Quit (Ping timeout: 480 seconds)
[17:29] * haomaiwa_ (~haomaiwan@119.6.74.59) has joined #ceph
[17:33] * shang (~ShangWu@210.242.156.234) Quit (Ping timeout: 480 seconds)
[17:34] * sagelap2 (~sage@38.122.20.226) Quit (Ping timeout: 480 seconds)
[17:38] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[17:39] * glowell1 (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) has joined #ceph
[17:40] * grepory1 (~Adium@12.236.17.3) has joined #ceph
[17:41] * chamings_ is now known as chamings
[17:42] * grepory (~Adium@12.236.17.3) Quit (Ping timeout: 480 seconds)
[17:42] * jeff-YF (~jeffyf@216.14.83.26) Quit (Quit: jeff-YF)
[17:43] * ScOut3R_ (~ScOut3R@catv-89-133-17-71.catv.broadband.hu) Quit (Ping timeout: 480 seconds)
[17:45] * sprachgenerator (~sprachgen@130.202.135.204) has joined #ceph
[17:45] * sagelap (~sage@2607:f298:a:607:f836:7a76:e39:2411) has joined #ceph
[17:45] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[17:46] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[17:52] * Vjarjadian (~IceChat77@176.254.37.210) Quit (Quit: The early bird may get the worm, but the second mouse gets the cheese)
[17:52] * bandrus (~Adium@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[17:54] * mdjp (~mdjp@gateb.kw.bbc.co.uk) has joined #ceph
[17:56] * mdjp (~mdjp@gateb.kw.bbc.co.uk) Quit (Remote host closed the connection)
[17:56] <loicd> :-D
[17:56] * xarses (~andreww@c-71-202-167-197.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[17:58] * mschiff (~mschiff@p4FD7F716.dip0.t-ipconnect.de) Quit (Remote host closed the connection)
[18:00] * shang (~ShangWu@210.242.156.234) has joined #ceph
[18:01] * mschiff (~mschiff@p4FD7F716.dip0.t-ipconnect.de) has joined #ceph
[18:04] <sagewk> nhm: fixed #, yeah. the mira/burnupi popped up immediately because the workers were idle
[18:04] * mschiff (~mschiff@p4FD7F716.dip0.t-ipconnect.de) Quit (Remote host closed the connection)
[18:05] <sagewk> loicd: no such thing as overkill :)
[18:06] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) has joined #ceph
[18:06] <sagewk> loicd: well, do you mean 32gb on the host or the guest?
[18:06] <sagewk> sounds like overkill for the guest..
[18:07] * shang (~ShangWu@210.242.156.234) Quit (Remote host closed the connection)
[18:08] * shang (~ShangWu@42-70-242-57.dynamic-ip.hinet.net) has joined #ceph
[18:08] * ircolle (~Adium@c-67-165-237-235.hsd1.co.comcast.net) has joined #ceph
[18:10] * xarses (~andreww@204.11.231.50.static.etheric.net) has joined #ceph
[18:13] * t0rn (~ssullivan@2607:fad0:32:a02:d227:88ff:fe02:9896) Quit (Remote host closed the connection)
[18:14] * BManojlovic (~steki@91.195.39.5) Quit (Quit: Ja odoh a vi sta 'ocete...)
[18:16] * jeff-YF (~jeffyf@216.14.83.26) has joined #ceph
[18:16] * mattt (~mattt@92.52.76.140) Quit (Read error: Connection reset by peer)
[18:16] * mjeanson (~mjeanson@00012705.user.oftc.net) Quit (Quit: No Ping reply in 180 seconds.)
[18:16] * mjeanson (~mjeanson@bell.multivax.ca) has joined #ceph
[18:18] * jeff-YF (~jeffyf@216.14.83.26) Quit ()
[18:21] <sagewk> yehudasa: yehuda_hm: ping
[18:26] * haomaiwa_ (~haomaiwan@119.6.74.59) Quit (Remote host closed the connection)
[18:26] * doxavore (~doug@99-7-52-88.lightspeed.rcsntx.sbcglobal.net) has joined #ceph
[18:26] * The_Bishop__ (~bishop@f055059070.adsl.alicedsl.de) has joined #ceph
[18:33] <jnq> is it possible to rebuild the header for a block image?
[18:33] * shang (~ShangWu@42-70-242-57.dynamic-ip.hinet.net) Quit (Read error: Connection reset by peer)
[18:33] * The_Bishop_ (~bishop@g227137157.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[18:33] * wogri_risc (~wogri_ris@85.233.124.80) has left #ceph
[18:39] * mxmln (~maximilia@212.79.49.65) has joined #ceph
[18:39] <mxmln> need urgent help
[18:43] * freedomhui (~freedomhu@117.79.232.248) Quit (Quit: Leaving...)
[18:43] * rendar (~s@host176-181-dynamic.10-87-r.retail.telecomitalia.it) Quit (Ping timeout: 480 seconds)
[18:45] <jmlowe> mxmln: with what?
[18:45] * odyssey4me (~odyssey4m@165.233.71.2) Quit (Ping timeout: 480 seconds)
[18:49] <jnq> jmlowe: do you know if it's possible to rebuild an rbd header?
[18:50] <jmlowe> jnq: just a user, you'd need to ask one of the inktank guys
[18:50] <jnq> jmlowe: np. i'm probably on the wrong timezone
[18:51] <jmlowe> should be prime time for inktank, most are on the west coast
[18:53] * alram (~alram@38.122.20.226) has joined #ceph
[18:55] * freedomhui (~freedomhu@117.79.232.247) has joined #ceph
[18:56] * houkouonchi-work (~linux@12.248.40.138) has joined #ceph
[18:57] <loicd> sagewk: I meant on the guest running teuthology. 16GB is enough ?
[18:57] <sagewk> oh, yeah probably!
[18:58] <loicd> ok cool
[19:00] * JustEra (~JustEra@ALille-555-1-86-217.w90-7.abo.wanadoo.fr) has joined #ceph
[19:00] * jeff-YF (~jeffyf@67.23.123.228) has joined #ceph
[19:00] * BManojlovic (~steki@fo-d-130.180.254.37.targo.rs) has joined #ceph
[19:01] <jnq> sagewk: do you know if it's possible to rebuild the header on an rbd image?
[19:04] * jeff-YF (~jeffyf@67.23.123.228) Quit ()
[19:06] * carif (~mcarifio@pool-96-233-32-122.bstnma.fios.verizon.net) has joined #ceph
[19:08] <loicd> zackc: speakng of VM size, if you had to provision VMs from OpenStack, would you provision them with the same flavor or tune the flavor depending on the job ? ( I can't see any easy way to do that except adding hints in the yaml )
[19:09] * smiley__ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) has joined #ceph
[19:10] * jeff-YF (~jeffyf@67.23.117.122) has joined #ceph
[19:10] * dpippenger (~riven@cpe-76-166-208-83.socal.res.rr.com) Quit (Quit: Leaving.)
[19:13] <sagewk> loicd: i would imagine there would be hints in the yaml to indicate what kind/size/flavor/whatever of vm to spin up
[19:15] * smiley_ (~smiley@pool-173-73-0-53.washdc.fios.verizon.net) Quit (Ping timeout: 480 seconds)
[19:15] * freedomhui (~freedomhu@117.79.232.247) Quit (Quit: Leaving...)
[19:17] <sagewk> zackc: maybe a generic wrapper 'adjust-ulimit <flag> <value>' ? then it is a single generic wrapper
[19:20] * shang (~ShangWu@122-116-16-162.HINET-IP.hinet.net) has joined #ceph
[19:21] * grepory1 (~Adium@12.236.17.3) Quit (Quit: Leaving.)
[19:21] * markbby (~Adium@168.94.245.1) Quit (Remote host closed the connection)
[19:22] * mschiff (~mschiff@85.182.236.82) has joined #ceph
[19:24] <zackc> sagewk: i was thinking that, yeah
[19:26] * grepory (~Adium@12.236.17.3) has joined #ceph
[19:29] * jbd_ (~jbd_@2001:41d0:52:a00::77) has left #ceph
[19:30] * sjm (~sjm@2607:f298:a:607:5cf3:56bd:3bb3:65c3) has joined #ceph
[19:31] <mxmln> our cluster is degraded may be some one can help: HEALTH_WARN 864 pgs backfill; 24 pgs degraded; 4 pgs down; 46 pgs incomplete; 33 pgs recovering; 892 pgs recovery_wait; 46 pgs stuck inactive; 1835 pgs stuck unclean; recovery 114146/480694 degraded (23.746%); 5867/206790 unfound (2.837%)
[19:31] <mxmln> http://pastebin.com/F0PY1wNn
[19:32] * madkiss (~madkiss@tmo-111-117.customers.d1-online.com) has joined #ceph
[19:32] <mxmln> recovering process has stopped at this state
[19:33] * gaveen (~gaveen@175.157.143.100) has joined #ceph
[19:34] * JustEra (~JustEra@ALille-555-1-86-217.w90-7.abo.wanadoo.fr) Quit (Quit: This computer has gone to sleep)
[19:36] <mxmln> health details points the problem always to the same 2 osd's (osd.0 and osd.1); ist it ok to take the both osd's down to continue the recovery??
[19:38] * sjm (~sjm@2607:f298:a:607:5cf3:56bd:3bb3:65c3) Quit (Remote host closed the connection)
[19:39] <Gugge-47527> if you have size=3 on the pool, yes you can take two osd's offline :)
[19:39] * sjm (~sjm@38.122.20.226) has joined #ceph
[19:40] * sjm (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[19:41] <mxmln> have rep size 2
[19:41] * sleinen1 (~Adium@2001:620:0:46:b41c:e261:23e9:e17e) has joined #ceph
[19:41] * sjm (~sjm@38.122.20.226) has joined #ceph
[19:41] <Gugge-47527> mxmln: then dont take two osds offline :)
[19:43] * sjm (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[19:43] * angdraug (~angdraug@204.11.231.50.static.etheric.net) has joined #ceph
[19:43] <mxmln> does it help to take down one by one?
[19:43] * sjm (~sjm@38.122.20.226) has joined #ceph
[19:44] <Gugge-47527> take one down, when the cluster is okay, take the other down
[19:44] <mxmln> taking one down and so that recovery continues and after next recovery stop taking another down
[19:45] * sjm (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[19:45] * sjm (~sjm@2607:f298:a:607:ac09:58a8:8c00:bffe) has joined #ceph
[19:48] * sleinen (~Adium@2001:620:0:25:e081:2938:5483:bdc3) Quit (Ping timeout: 480 seconds)
[19:53] * rturk-away is now known as rturk
[19:54] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) has joined #ceph
[19:56] * sjm (~sjm@2607:f298:a:607:ac09:58a8:8c00:bffe) Quit (Remote host closed the connection)
[19:56] * alfredodeza is now known as alfredo|noms
[19:58] * sjm (~sjm@38.122.20.226) has joined #ceph
[19:59] * markbby (~Adium@168.94.245.1) has joined #ceph
[20:00] * dpippenger (~riven@tenant.pas.idealab.com) has joined #ceph
[20:04] * ScOut3R (~scout3r@BC2484D1.dsl.pool.telekom.hu) has joined #ceph
[20:05] * grepory (~Adium@12.236.17.3) Quit (Quit: Leaving.)
[20:08] * grepory (~Adium@12.236.17.3) has joined #ceph
[20:09] * ScOut3R (~scout3r@BC2484D1.dsl.pool.telekom.hu) Quit (Remote host closed the connection)
[20:09] * ScOut3R (~scout3r@BC2484D1.dsl.pool.telekom.hu) has joined #ceph
[20:10] * loNely_gRavE (loNely_gRa@117.201.202.175) has joined #ceph
[20:11] * loNely_gRavE (loNely_gRa@117.201.202.175) Quit ()
[20:26] * grepory1 (~Adium@2600:1010:b01c:7cc1:38db:f9dc:7667:3a38) has joined #ceph
[20:32] * grepory (~Adium@12.236.17.3) Quit (Ping timeout: 480 seconds)
[20:36] * grepory1 (~Adium@2600:1010:b01c:7cc1:38db:f9dc:7667:3a38) Quit (Read error: Connection reset by peer)
[20:41] * JustEra (~JustEra@ALille-555-1-86-217.w90-7.abo.wanadoo.fr) has joined #ceph
[20:42] * sjm (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[20:43] * grepory (~Adium@2600:1010:b01c:7cc1:2805:35bf:53bd:6c71) has joined #ceph
[20:47] * rturk is now known as rturk-away
[20:51] * allsystemsarego (~allsystem@188.25.134.187) Quit (Quit: Leaving)
[20:53] <mtanski> I'm trying to debug a bug in the cephfs directory handling code.
[20:53] * Cube (~Cube@cpe-76-95-217-129.socal.res.rr.com) Quit (Quit: Leaving.)
[20:53] <mtanski> If have a large directory structure (say a 5k) subdirectories when I run: find -type f | xargs -P 10 cat | pv > /dev/null but after a while it complains of missing files
[20:53] <mtanski> no one else using the mount point
[20:54] <mtanski> and when i do ls, the number if directories becomes much lower
[20:54] <mtanski> when i touch the directory everything is okay again
[20:54] <mtanski> I'm not sure which corner of the cephfs code to look for this
[20:54] * JustEra (~JustEra@ALille-555-1-86-217.w90-7.abo.wanadoo.fr) Quit (Quit: This computer has gone to sleep)
[20:54] * alfredo|noms is now known as alfredodeza
[20:57] * sjm (~sjm@2607:f298:a:607:25bd:d6d7:544:b16d) has joined #ceph
[20:57] <mtanski> this happens without dcache and without fscache
[20:57] <mtanski> as well
[21:01] * grepory (~Adium@2600:1010:b01c:7cc1:2805:35bf:53bd:6c71) Quit (Quit: Leaving.)
[21:02] * jluis (~joao@89-181-152-211.net.novis.pt) has joined #ceph
[21:02] * ChanServ sets mode +o jluis
[21:05] <loicd> sagewk: https://github.com/ceph/ceph/pull/518 is now ready to merge ( the rebase was a trivial conflict in Makefile.am ). It ran make check successfully.
[21:07] * shang (~ShangWu@122-116-16-162.HINET-IP.hinet.net) Quit (Ping timeout: 480 seconds)
[21:08] * joao (~joao@89.181.152.211) Quit (Ping timeout: 480 seconds)
[21:08] * rturk-away is now known as rturk
[21:10] * mikedawson (~chatzilla@23-25-46-97-static.hfc.comcastbusiness.net) has joined #ceph
[21:16] * Cube (~Cube@12.248.40.138) has joined #ceph
[21:24] * nwat (~nwat@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Read error: Operation timed out)
[21:24] * lx0 (~aoliva@lxo.user.oftc.net) Quit (Remote host closed the connection)
[21:25] * lx0 (~aoliva@lxo.user.oftc.net) has joined #ceph
[21:29] * PITon (~pavel@195.182.195.107) Quit (Read error: Connection reset by peer)
[21:29] * PITon (~pavel@195.182.195.107) has joined #ceph
[21:33] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) Quit (Ping timeout: 480 seconds)
[21:39] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) has joined #ceph
[21:41] * nwat (~nwat@eduroam-237-79.ucsc.edu) has joined #ceph
[21:46] * diegows (~diegows@200.68.116.185) Quit (Read error: Connection reset by peer)
[21:47] * LPG (~LPG@c-76-104-197-224.hsd1.wa.comcast.net) Quit (Ping timeout: 480 seconds)
[21:48] * lx0 is now known as lxo
[21:48] * Vjarjadian (~IceChat77@176.254.37.210) has joined #ceph
[21:49] * rturk is now known as rturk-away
[21:52] <zackc> sagewk: so, only 'ceph-osd' needs -n, not 'ceph osd' ?
[21:56] * joao (~joao@89-181-152-211.net.novis.pt) has joined #ceph
[21:56] * ChanServ sets mode +o joao
[21:58] * davidzlap (~Adium@ip68-5-239-214.oc.oc.cox.net) Quit (Quit: Leaving.)
[22:00] * grepory (~Adium@2600:1010:b019:3194:70c3:341c:f109:b037) has joined #ceph
[22:02] * jluis (~joao@89-181-152-211.net.novis.pt) Quit (Ping timeout: 480 seconds)
[22:02] * rturk-away is now known as rturk
[22:06] * sprachgenerator (~sprachgen@130.202.135.204) Quit (Read error: Connection reset by peer)
[22:06] <mikedawson> Is there a way to differentiate 'ceph -w' type clients from rbd clients? Both get the rules under [client], is there something like [client.rbd] or something?
[22:07] * grepory (~Adium@2600:1010:b019:3194:70c3:341c:f109:b037) Quit (Quit: Leaving.)
[22:07] <dmick> mikedawson: you can set the client name; ceph command line args -n/--name or -i/--id
[22:07] <dmick> with libvirt I think that goes in CEPH_ARGS IIRC
[22:07] * sprachgenerator (~sprachgen@130.202.135.204) has joined #ceph
[22:08] <dmick> I guess --id is the relevant one for 'client' programs, because it'll always have a type of 'client'
[22:08] <dmick> ceph-rest-api uses client.restapi, for example, by default
[22:10] <mikedawson> dmick: seems that changing --id changes the authentication key it tries to pickup
[22:11] <mikedawson> dmick: and name looks like this-> http://pastebin.com/raw.php?i=1yrPfSMw
[22:11] <dmick> yes. the type.id 'name' is used for authentication as well as configuration
[22:11] <dmick> (and yeah, you can't really use --name except in the daemons)
[22:11] <mikedawson> ok
[22:13] <sagewk> alfredodeza: on 6017: use 'ceph daemon mon.`hostname` mon-status'
[22:14] <sagewk> that goes through the admin socket instead of over the network, and skips the normal auth stuff and doesn't require the mons be quorum.
[22:14] * The_Bishop__ (~bishop@f055059070.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[22:14] <dmick> (of course you need to run that on the host that has the mon process)
[22:16] <alfredodeza> issue 6017
[22:16] <kraken> alfredodeza might be talking about: http://tracker.ceph.com/issues/6017 [ceph-deploy mon create: create on all mons in ceph.conf + then do gatherkeys if no mons are explicitly listed]
[22:18] <sagewk> zackc: what about just http://fpaste.org/37434/41226413/ ? then the ... adjust-ulimits ... becomes ... adjust-ulimit c unlimited adjust-ulimit n 16384 ... (with that last bit only when running ceph-osd)
[22:20] * peetaur (~peter@CPEbc1401e60493-CMbc1401e60490.cpe.net.cable.rogers.com) Quit (Ping timeout: 480 seconds)
[22:21] * sjm (~sjm@2607:f298:a:607:25bd:d6d7:544:b16d) Quit (Remote host closed the connection)
[22:21] * sjm (~sjm@38.122.20.226) has joined #ceph
[22:21] <zackc> sagewk: is that better somehow?
[22:23] <sagewk> puts the ceph-osd special-case in the caller (who knows what they want) instead of buried in the helper
[22:23] * BManojlovic (~steki@fo-d-130.180.254.37.targo.rs) Quit (Quit: Ja odoh a vi sta 'ocete...)
[22:23] <sagewk> it does make the command line longer, but if adjust-ulimit is in the path it seems ok to me
[22:23] <sagewk> and makes it easy to adjust other things in other ways later
[22:23] * The_Bishop__ (~bishop@f055059070.adsl.alicedsl.de) has joined #ceph
[22:25] <zackc> well, it's not currently in the path
[22:25] <zackc> i haven't found a way to set $PATH with paramiko
[22:26] <zackc> also this patch would be far easier to cherry-pick into other teuthology branches
[22:26] * peetaur (~peter@CPEbc1401e60493-CMbc1401e60490.cpe.net.cable.rogers.com) has joined #ceph
[22:26] <zackc> so that we could actually remove something from chef
[22:33] <sagewk> zackc: hmm. yeah i think it's perfect for the cherry-pick
[22:34] <sagewk> but all things being equal i prefer the generic helper for moving forward
[22:35] * chiluk (~chiluk@cpe-70-124-70-187.austin.res.rr.com) has joined #ceph
[22:37] <mtanski> Sage, I spoke to David Howells. He fix up one two of the fscache packages (the code that checks object validity)
[22:38] <mtanski> He was asking if you can push that + the ceph fscache patches via the Ceph. I've rebase all the changes and have all the proper sign offs
[22:41] <sagewk> sure
[22:41] * sjustlaptop (~sam@24-205-35-233.dhcp.gldl.ca.charter.com) has joined #ceph
[22:41] <sagewk> does that include those first two adding the revalidate interface? i didn't see those in his fscache branch
[22:42] <sagewk> mtanski: want to send a ref to a git branch I can pull from?
[22:43] <sagewk> mtanski: just to clarify: *everything* shoudl go through the ceph tree? we're not waiting on anything from him directly?
[22:43] <mtanski> And on a unrelated note, how would I go about debuting an issue in the dir code when Ceph loses dirents on a large directory
[22:43] <mtanski> Yeah, I've spoke to him on irc like 2 hours ago
[22:45] <sagewk> mtanski: i'm hopeful that will get resolved by the d_prune patch from yanzheng that viro just sent to linus
[22:46] * terje__ (~root@135.109.220.9) has joined #ceph
[22:47] * terje__ is now known as terje-
[22:47] <terje-> hi, I notice that in ceph -s/-w output I get a very nice XXop/s
[22:47] <terje-> is there a way to get an average of that over some amount of time?
[22:48] <terje-> like, in the last day we saw YYop/s on average
[22:51] * Midnightmyth_ (~quassel@93-167-84-102-static.dk.customer.tdc.net) Quit (Ping timeout: 480 seconds)
[22:53] <mtanski> I'll send a new email with the whole patchset, and CC everybody, David can give his blessing for it to be router via Ceph on the mailing list and everything should be kosher.
[22:53] <mtanski> Sage, do you have a link to the change from viro?
[22:55] * tryggvil (~tryggvil@178.19.53.254) Quit (Quit: tryggvil)
[23:00] * markbby (~Adium@168.94.245.1) Quit (Quit: Leaving.)
[23:00] * sjm (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[23:01] * sjm (~sjm@2607:f298:a:607:48f5:801d:ab08:7968) has joined #ceph
[23:03] * sjm (~sjm@2607:f298:a:607:48f5:801d:ab08:7968) Quit (Remote host closed the connection)
[23:03] * sjm (~sjm@38.122.20.226) has joined #ceph
[23:05] * sjm (~sjm@38.122.20.226) Quit (Remote host closed the connection)
[23:05] * sjm (~sjm@2607:f298:a:607:c11c:3245:fbca:7d92) has joined #ceph
[23:08] * nhorman (~nhorman@hmsreliant.think-freely.org) Quit (Quit: Leaving)
[23:12] * jks (~jks@3e6b5724.rev.stofanet.dk) has joined #ceph
[23:12] * The_Bishop__ (~bishop@f055059070.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[23:22] * mikedawson (~chatzilla@23-25-46-97-static.hfc.comcastbusiness.net) Quit (Ping timeout: 480 seconds)
[23:26] * nhm (~nhm@184-97-244-237.mpls.qwest.net) Quit (Read error: Operation timed out)
[23:28] * dmsimard (~Adium@108.163.152.2) Quit (Ping timeout: 480 seconds)
[23:29] * madkiss (~madkiss@tmo-111-117.customers.d1-online.com) Quit (Quit: Leaving.)
[23:31] * nhm (~nhm@mfe2836d0.tmodns.net) has joined #ceph
[23:31] * KindTwo (~KindOne@198.14.194.32) has joined #ceph
[23:32] * KindOne (~KindOne@0001a7db.user.oftc.net) Quit (Ping timeout: 480 seconds)
[23:32] * KindTwo is now known as KindOne
[23:33] * mschiff (~mschiff@85.182.236.82) Quit (Remote host closed the connection)
[23:37] * mschiff (~mschiff@85.182.236.82) has joined #ceph
[23:39] <nigwil> one of my OSDs got too full, and now has this: _detect_fs unable to create /var/lib/ceph/osd/ceph-10/xattr_test: (28) No space left on device
[23:39] * lordinvader (~lordinvad@14.139.82.6) has joined #ceph
[23:39] <nigwil> is there any easy fix or should I recreate the OSD?
[23:44] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) Quit (Remote host closed the connection)
[23:45] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) has joined #ceph
[23:45] * thomnico_ (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) has joined #ceph
[23:46] * jcsp (~john@82-71-55-202.dsl.in-addr.zen.co.uk) has joined #ceph
[23:47] * tryggvil (~tryggvil@17-80-126-149.ftth.simafelagid.is) has joined #ceph
[23:47] * carif (~mcarifio@pool-96-233-32-122.bstnma.fios.verizon.net) Quit (Quit: Ex-Chat)
[23:47] * carif (~mcarifio@pool-96-233-32-122.bstnma.fios.verizon.net) has joined #ceph
[23:50] * mschiff (~mschiff@85.182.236.82) Quit (Remote host closed the connection)
[23:55] * thomnico_ (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) Quit (Quit: Ex-Chat)
[23:55] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) Quit (Quit: Ex-Chat)
[23:55] * thomnico (~thomnico@2a01:e35:8b41:120:e051:2a8e:a6fe:ba3d) has joined #ceph
[23:57] * vata (~vata@2607:fad8:4:6:ad7b:c365:d5eb:b7b2) Quit (Quit: Leaving.)

These logs were automatically created by CephLogBot on irc.oftc.net using the Java IRC LogBot.