#ceph IRC Log


IRC Log for 2011-02-23

Timestamps are in GMT/BST.

[0:18] * cmccabe (~cmccabe@ has joined #ceph
[0:35] * verwilst (~verwilst@dD576FAAE.access.telenet.be) Quit (Quit: Ex-Chat)
[1:39] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[1:58] * Tv notes 7-machine autotest runs are starting to bring out its constant overheads..
[2:03] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[2:10] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[3:36] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[4:32] * cmccabe (~cmccabe@ Quit (Quit: Leaving.)
[5:37] * cclien (~cclien@ec2-175-41-146-71.ap-southeast-1.compute.amazonaws.com) Quit (charon.oftc.net kilo.oftc.net)
[5:37] * morse (~morse@supercomputing.univpm.it) Quit (charon.oftc.net kilo.oftc.net)
[5:37] * hijacker (~hijacker@ Quit (charon.oftc.net kilo.oftc.net)
[5:37] * alexxy (~alexxy@ Quit (charon.oftc.net kilo.oftc.net)
[5:37] * andret (~andre@pcandre.nine.ch) Quit (charon.oftc.net kilo.oftc.net)
[5:37] * stingray (~stingray@stingr.net) Quit (charon.oftc.net kilo.oftc.net)
[5:37] * cclien (~cclien@ec2-175-41-146-71.ap-southeast-1.compute.amazonaws.com) has joined #ceph
[5:37] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[5:37] * hijacker (~hijacker@ has joined #ceph
[5:37] * alexxy (~alexxy@ has joined #ceph
[5:37] * andret (~andre@pcandre.nine.ch) has joined #ceph
[5:37] * stingray (~stingray@stingr.net) has joined #ceph
[7:41] * prometheanfire (~mthode@mx1.mthode.org) Quit (Ping timeout: 480 seconds)
[7:42] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[7:55] * Juul (~Juul@nat-sonicnet.noisebridge.net) has joined #ceph
[8:21] * prometheanfire (~mthode@mx1.mthode.org) has joined #ceph
[9:10] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[9:59] * Yoric (~David@ has joined #ceph
[10:26] * MK_FG (~MK_FG@ Quit (Quit: o//)
[10:29] * MK_FG (~MK_FG@ has joined #ceph
[10:33] * allsystemsarego (~allsystem@ has joined #ceph
[11:29] * verwilst (~verwilst@router.begen1.office.netnoc.eu) has joined #ceph
[12:17] * yx (~yx@19NAAE029.tor-irc.dnsbl.oftc.net) Quit (Ping timeout: 480 seconds)
[12:25] * yx (~yx@28IAACQ7U.tor-irc.dnsbl.oftc.net) has joined #ceph
[12:54] * Yoric_ (~David@ has joined #ceph
[12:54] * Yoric (~David@ Quit (Read error: Connection reset by peer)
[12:54] * Yoric_ is now known as Yoric
[12:57] * Juul (~Juul@nat-sonicnet.noisebridge.net) Quit (Read error: Operation timed out)
[13:19] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (Ping timeout: 480 seconds)
[13:19] * tjikkun (~tjikkun@195-240-187-63.ip.telfort.nl) has joined #ceph
[14:08] * alexxy[home] (~alexxy@masq246.gtn.ru) has joined #ceph
[14:12] * alexxy (~alexxy@ Quit (Ping timeout: 480 seconds)
[14:17] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[14:35] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[16:30] * squig (~bendeluca@soho-94-143-249-50.sohonet.co.uk) Quit (Quit: Ex-Chat)
[16:48] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Remote host closed the connection)
[16:51] <johnl> hi all
[17:03] <wido> hi johnl
[17:17] <wido> sagewk: Looks like some PG's are blocking on my cluster. "8240 pgs: 8176 active+clean, 64 active+scrubbing+repair"
[17:17] <wido> I tried to find out what a "repair" would do, but that is blocking now for about 6 hours
[17:30] * verwilst (~verwilst@router.begen1.office.netnoc.eu) Quit (Quit: Ex-Chat)
[17:42] * gregaf (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[18:15] * Tv (~Tv|work@ip-66-33-206-8.dreamhost.com) has joined #ceph
[18:19] * cmccabe (~cmccabe@c-24-23-253-6.hsd1.ca.comcast.net) has joined #ceph
[18:38] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[18:44] * ghaskins (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Ping timeout: 480 seconds)
[18:49] * ghaskins (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[18:54] * Yoric (~David@ Quit (Quit: Yoric)
[19:05] * bchrisman (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[19:06] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[19:07] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:13] <Tv> how do i use cauthtool to manipulate caps of a specific key in the keyring? it touches client.admin by default..
[19:13] <greglap> there's some option to specify the user
[19:13] <greglap> same option as when you're creating a new user, I think
[19:13] <Tv> const char *name = g_conf.name;
[19:14] <Tv> i don't see that being overwritten, except for --gen-key etc
[19:14] <greglap> no, I don't think it's that one — there's a user name or user id or something (distinct from auid)
[19:14] <Tv> that gets shuffled into EntityName ename, and really looks like the thing that matters
[19:15] <Tv> e.g. keyring.add(ename, eauth);
[19:15] <Tv> keyring.set_caps(ename, caps);
[19:16] <Tv> i don't see where argv is used to set that though, even for --gen-key
[19:16] <greglap> I thought the man page for cauthtool actually had this kind of documentation
[19:16] <greglap> did you check that?
[19:16] <Tv> haha
[19:16] <Tv> greglap: good one!
[19:17] <greglap> hmm, okay
[19:17] <Tv> greglap: in case you haven't noticed, ceph usage strings and man pages are mostly unrelated to reality, when you go outside the few options that are documented
[19:17] <greglap> I just thought I remembered documenting this at some point :)
[19:17] <greglap> I do see this though:
[19:17] <Tv> oh wow it's usage summary says --nanme
[19:17] <Tv> --name
[19:18] <Tv> so cauthtool itself is out of sync with that
[19:18] <greglap> "To associate some capabilities with the key (namely, the ability to mount a Ceph filesystem):"
[19:18] <greglap> "cauthtool -n client.foo —cap mds 'allow' —cap osd 'allow rw pool=data' —cap mon 'allow r' keyring
[19:19] <Tv> yeah cauthtool itself not having that threw me ofdf
[19:19] <Tv> have i mentioned i hate common_init and non-explicit shared argv parsing?
[19:19] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[19:19] <greglap> so I guess it is just the name after all
[19:19] <Tv> yeah seems so
[19:20] <Tv> funny thing also, if you use --caps on a key that doesn't exist, you get this: key = AAAAAAAAAAAAAAAA
[19:20] <Tv> not very helpful ;)
[19:20] <greglap> is that the computer shrieking in pain?
[19:25] <wido> I know that seeing "connect claims to be", "wrong node!" is not that bad, but my OSD's keep logging those messages, is that normal?
[19:25] <wido> they log if every 30 seconds
[19:25] * MK_FG (~MK_FG@ Quit (Quit: o//)
[19:26] * Meths_ (rift@ has joined #ceph
[19:32] * Meths (rift@ Quit (Ping timeout: 480 seconds)
[19:39] <cmccabe> the last 5 minutes of that meeting sounded autotuned
[19:42] <Tv> more like the autotune equivalent of beatboxing
[19:44] <Tv> sagewk: but yeah, fyi your upstream bandwidth was not enough to handle sending video at all, and even voice was droppng
[19:44] <greglap> it was doing okay for the first half — low fps but stable
[19:44] <Tv> yeah
[19:45] <Tv> the first half kinda auto-disabled video in favor of audio
[19:48] * sage (~sage@dsl092-035-022.lax1.dsl.speakeasy.net) has left #ceph
[19:49] <Tv> so.. i have "ceph auth add" hanging, and 3 cmons all eating ~20% cpu
[19:49] <Tv> strace claims they all sit in a futex() call, but i do believe that is a lie
[19:49] <Tv> (due to threads)
[19:50] * greglap1 (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:50] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Read error: Connection reset by peer)
[19:51] <Tv> hmm ceph health hangs too
[19:51] <cmccabe> turn on messenger debug?
[19:51] <Tv> i see action on the tcp port
[19:54] <Tv> 2011-02-23 10:54:01.914909 7f780a8b4700 mon.0@0(starting) e1 ms_verify_authorizer cephx enabled, but no authorizer (required for mon)
[19:54] <Tv> i screwed up the keys, it seems
[19:55] <Tv> ooh the mons hate each other, it seems
[20:08] * greglap1 (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Ping timeout: 480 seconds)
[20:11] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[20:11] <Tv> at what time does fsid get assigned?
[20:12] <Tv> ah in monmap, not in osdmap
[20:12] <Tv> i was confused by seeing all 0's in osdmaptool output
[20:14] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[20:14] * fzylogic (~fzylogic@ has joined #ceph
[20:24] * gregaf (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[20:25] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[20:37] <Tv> woo 7-node autotest worked!
[20:37] <Tv> it knows how to build an arbitrary cluster, just list the what roles go on each host and it'll set it up
[20:38] <sjust> :)
[20:49] * alexxy[home] (~alexxy@masq246.gtn.ru) Quit (Quit: No Ping reply in 180 seconds.)
[20:49] * alexxy (~alexxy@ has joined #ceph
[20:52] * MK_FG (~MK_FG@ has joined #ceph
[21:20] * DJLee (82d8d198@ircip2.mibbit.com) has joined #ceph
[21:22] * WesleyS (~WesleyS@ has joined #ceph
[21:24] * bchrisman (~Adium@70-35-37-146.static.wiline.com) Quit (Quit: Leaving.)
[21:28] <wido> sagewk: Did you get the chance to look at the stalling recovery? To me it seems a couple of PG's are blocking
[22:24] * Guest1881 (quasselcor@bas11-montreal02-1128535815.dsl.bell.ca) Quit (Remote host closed the connection)
[22:24] * bbigras (quasselcor@bas11-montreal02-1128535815.dsl.bell.ca) has joined #ceph
[22:25] * bbigras is now known as Guest2417
[22:25] * WesleyS (~WesleyS@ has left #ceph
[22:40] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[23:03] * ghaskins_mobile (~ghaskins_@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Quit: This computer has gone to sleep)
[23:58] <Tv> OPTION(client_mountpoint, 'r', OPT_STR, "/"),
[23:58] <Tv> oh that's bullshit

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