#ceph IRC Log

Index

IRC Log for 2012-09-28

Timestamps are in GMT/BST.

[0:09] * mtk (~mtk@ool-44c35bb4.dyn.optonline.net) Quit (Remote host closed the connection)
[0:16] * Cube (~Adium@2607:f298:a:697:ca2a:14ff:fe16:4a67) Quit (Quit: Leaving.)
[0:16] * Cube (~Adium@38.122.20.226) has joined #ceph
[0:19] * jefferai (~quassel@quassel.jefferai.org) Quit (Remote host closed the connection)
[0:22] * mtk (~mtk@ool-44c35bb4.dyn.optonline.net) has joined #ceph
[0:45] * Cube (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[0:46] * jjgalvez (~jjgalvez@38.122.20.226) Quit (Quit: Leaving.)
[0:48] * KevinPerks1 (~Adium@2607:f298:a:607:fdcc:55a8:c0c4:82ac) Quit (Quit: Leaving.)
[0:49] * KevinPerks (~Adium@2607:f298:a:607:2d71:a473:d953:e8d) has joined #ceph
[0:54] * jefferai (~quassel@quassel.jefferai.org) has joined #ceph
[0:56] * KevinPerks (~Adium@2607:f298:a:607:2d71:a473:d953:e8d) Quit (Quit: Leaving.)
[0:58] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[0:59] * jmlowe (~Adium@adsl-99-124-128-85.dsl.ipltin.sbcglobal.net) Quit (Quit: Leaving.)
[0:59] * KevinPerks (~Adium@38.122.20.226) Quit (Read error: Connection reset by peer)
[0:59] * KevinPerks (~Adium@2607:f298:a:607:38ba:91a1:4def:16da) has joined #ceph
[1:00] * adjohn (~adjohn@108-225-130-229.lightspeed.sntcca.sbcglobal.net) Quit (Quit: adjohn)
[1:11] <sjust> mikeryan: the simplest thing is probably to collapse the backfill reservation thing into a more general recovery reservation
[1:12] <sjust> mikeryan: currently, we maintain an OSD wide recovery_ops counter (see OSD::start_recovery_op, etc), that could become a per pg concept since we'll be limiting the number of pgs recovering
[1:15] <mikeryan> i see
[1:15] <mikeryan> so a backfill will just be sequenced like any other recovery op
[1:15] <sjust> that's essentially how it already works
[1:16] <sjust> each scan or push is a recovery o
[1:16] <sjust> *p
[1:17] <sjust> PG::start_recovery_op currently calls up into OSD::start_recovery_op to increment recovery_ops_active
[1:17] <sjust> instead, it could stop with it's local counter and use that in start_recovery_ops to make decisions
[1:18] <sjust> it removes an annoying osd state dependency as a bonus
[1:24] <mikeryan> that's a blessing and a curse
[1:24] <mikeryan> now we no longer limit the total number of ops per OSD
[1:24] <sjust> we do
[1:24] <sjust> outgoing limit = max_outgoing_reservations * max_per_pg_ops
[1:25] <sjust> incoming limit = max_incoming_reservations * max_per_pg_ops
[1:25] <gregaf1> but it means you can't decide to promote recovery on a specific PG over another PG
[1:25] <gregaf1> by allocating it more of the in-flight ops
[1:25] <gregaf1> (just a counterpoint to consider)
[1:26] <sjust> gregaf1: true, the primary case there is triggering a push immediately due to client op on degraded object, those will use a seperate mechanism and are implicitly limited by client op limitations
[1:26] <sjust> since the client op has to wait for the push to complete
[1:26] <sjust> that should be the only case where we want to expedite a particular push
[1:27] <gregaf1> yeah, I don't think it's a big issue right now; just considering a bit more intelligence like wanting to prioritize more-degraded PGs, for which the local node is supposed to be primary, etc
[1:27] <sjust> we would choose which pg to grab a reservation for
[1:27] <sjust> rather than doing it at the per op level
[1:28] <gregaf1> and then that PG gets 1/4 of the overall recovery bandwidth, and you can't assign it half or whatever
[1:28] <gregaf1> not a dealbreaker!
[1:28] <sjust> yup
[1:28] <sjust> it's a cost, but not a big one
[1:29] <mikeryan> the benefit of the system we're discussing is that it's simple
[1:29] <mikeryan> both to implement and to understand
[1:29] <sjust> yeah
[1:29] <mikeryan> which hopefully means fewer bugs
[1:30] <sjust> besides, the actual throughput is limited by client vs recovery op throttling
[1:30] <sjust> nvm, my previous statement is not relevant
[1:30] <gregaf1> I just wanted somebody to consider that point and decide they don't care ;)
[1:30] <mikeryan> so noted
[1:32] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) Quit (Quit: Leseb)
[1:32] * scuttlemonkey (~scuttlemo@c-69-244-181-5.hsd1.mi.comcast.net) has joined #ceph
[1:33] * benpol (~benp@garage.reed.edu) has left #ceph
[1:41] * nhm (~nhm@174-20-32-79.mpls.qwest.net) has joined #ceph
[1:42] * nhm_ (~nhm@67-220-20-222.usiwireless.com) Quit (Ping timeout: 480 seconds)
[1:51] * yoshi (~yoshi@p37219-ipngn1701marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[1:51] * KevinPerks (~Adium@2607:f298:a:607:38ba:91a1:4def:16da) Quit (Read error: Connection reset by peer)
[1:51] * KevinPerks (~Adium@2607:f298:a:607:38ba:91a1:4def:16da) has joined #ceph
[1:59] * SvenDowideit (~SvenDowid@203-206-171-38.perm.iinet.net.au) Quit (Read error: Operation timed out)
[2:01] * SvenDowideit (~SvenDowid@203-206-171-38.perm.iinet.net.au) has joined #ceph
[2:02] * jmlowe (~Adium@c-71-201-31-207.hsd1.in.comcast.net) has joined #ceph
[2:02] * gaveen (~gaveen@112.135.156.122) Quit (Ping timeout: 480 seconds)
[2:11] * gaveen (~gaveen@112.135.148.69) has joined #ceph
[2:19] * KevinPerks1 (~Adium@38.122.20.226) has joined #ceph
[2:22] * KevinPerks (~Adium@2607:f298:a:607:38ba:91a1:4def:16da) Quit (Ping timeout: 480 seconds)
[2:24] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Ping timeout: 480 seconds)
[2:29] * sagelap1 (~sage@190.sub-70-197-146.myvzw.com) has joined #ceph
[2:30] * sagelap (~sage@38.122.20.226) Quit (Ping timeout: 480 seconds)
[2:31] * Tv_ (~tv@2607:f298:a:607:b899:20f7:e1bb:234c) Quit (Quit: Tv_)
[2:37] * KevinPerks1 (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[2:37] * sagelap1 (~sage@190.sub-70-197-146.myvzw.com) Quit (Ping timeout: 480 seconds)
[2:43] * adjohn (~adjohn@108-225-130-229.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[2:49] * sagelap (~sage@159.sub-70-197-150.myvzw.com) has joined #ceph
[2:50] * adjohn (~adjohn@108-225-130-229.lightspeed.sntcca.sbcglobal.net) Quit (Quit: adjohn)
[3:19] * joshd (~joshd@2607:f298:a:607:221:70ff:fe33:3fe3) Quit (Quit: Leaving.)
[3:23] * Karcaw (~evan@68-186-68-219.dhcp.knwc.wa.charter.com) Quit (Ping timeout: 480 seconds)
[3:28] * Cube (~Adium@cpe-76-95-223-199.socal.res.rr.com) has joined #ceph
[3:28] * sagelap (~sage@159.sub-70-197-150.myvzw.com) Quit (Ping timeout: 480 seconds)
[3:31] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[3:31] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[3:32] * Karcaw (~evan@68-186-68-219.dhcp.knwc.wa.charter.com) has joined #ceph
[3:37] * nhm_ (~nhm@67-220-20-222.usiwireless.com) has joined #ceph
[3:38] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[3:40] * nhm (~nhm@174-20-32-79.mpls.qwest.net) Quit (Ping timeout: 480 seconds)
[3:45] * KevinPerks (~Adium@38.122.20.226) Quit (Read error: Operation timed out)
[3:54] * chutzpah (~chutz@100.42.98.5) Quit (Quit: Leaving)
[4:38] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[4:46] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[4:50] * zodiak (~stef@CPE2cb05d3ebdcb-CM602ad07b9954.cpe.net.cable.rogers.com) Quit (Quit: Leaving)
[5:08] * maelfius (~mdrnstm@66.209.104.107) Quit (Quit: Leaving.)
[5:13] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[5:14] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[5:33] * bugfixer (~dchambers@42gis175.gulftel.com) Quit (Ping timeout: 480 seconds)
[5:37] * deepsa (~deepsa@122.172.16.106) Quit (Quit: Computer has gone to sleep.)
[5:38] * slang (~slang@38.122.20.226) Quit (Quit: slang)
[5:39] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[5:41] * deepsa (~deepsa@122.172.16.106) has joined #ceph
[5:47] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[5:51] * sagelap (~sage@76.89.177.113) has joined #ceph
[5:57] * jjgalvez (~jjgalvez@cpe-76-175-17-226.socal.res.rr.com) has joined #ceph
[6:11] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) has joined #ceph
[6:21] * deepsa (~deepsa@122.172.16.106) Quit (Remote host closed the connection)
[6:24] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[6:24] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[6:24] * deepsa (~deepsa@122.166.163.68) has joined #ceph
[6:26] * glowell (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[6:39] * KevinPerks (~Adium@2607:f298:a:607:1046:cf0d:2400:5d39) has joined #ceph
[6:40] * glowell (~Adium@68.170.71.123) has joined #ceph
[6:42] * deepsa_ (~deepsa@122.172.2.82) has joined #ceph
[6:44] * KevinPerks (~Adium@2607:f298:a:607:1046:cf0d:2400:5d39) Quit (Read error: Operation timed out)
[6:45] * deepsa (~deepsa@122.166.163.68) Quit (Ping timeout: 480 seconds)
[6:45] * deepsa_ is now known as deepsa
[6:55] * Ryan_Lane (~Adium@c-67-160-217-184.hsd1.ca.comcast.net) has joined #ceph
[7:00] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[7:01] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[7:03] * jjgalvez (~jjgalvez@cpe-76-175-17-226.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[7:15] * Ryan_Lane (~Adium@c-67-160-217-184.hsd1.ca.comcast.net) has left #ceph
[7:21] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[7:21] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[7:30] * SvenDowideit (~SvenDowid@203-206-171-38.perm.iinet.net.au) Quit (Ping timeout: 480 seconds)
[7:31] * SvenDowideit (~SvenDowid@203-206-171-38.perm.iinet.net.au) has joined #ceph
[7:39] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[7:47] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[7:49] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) Quit (Quit: Leaving.)
[7:58] * MikeMcClurg (~mike@cpc10-cmbg15-2-0-cust205.5-4.cable.virginmedia.com) has joined #ceph
[8:08] * glowell (~Adium@68.170.71.123) Quit (Ping timeout: 480 seconds)
[8:18] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[8:26] * Cube (~Adium@cpe-76-95-223-199.socal.res.rr.com) Quit (Quit: Leaving.)
[8:40] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[8:41] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[8:45] * KevinPerks (~Adium@38.122.20.226) Quit (Read error: Operation timed out)
[8:48] * EmilienM (~EmilienM@195-132-228-252.rev.numericable.fr) has joined #ceph
[9:02] * guerby (~guerby@nc10d-ipv6.tetaneutral.net) Quit (Ping timeout: 480 seconds)
[9:30] * loicd (~loic@178.20.50.225) has joined #ceph
[9:31] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[9:40] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[9:43] * BManojlovic (~steki@87.110.183.173) has joined #ceph
[9:48] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[9:48] * MikeMcClurg (~mike@cpc10-cmbg15-2-0-cust205.5-4.cable.virginmedia.com) Quit (Quit: Leaving.)
[9:59] * loicd (~loic@178.20.50.225) Quit (Ping timeout: 480 seconds)
[10:03] * ninkotech_ (~duplo@89.177.137.231) has joined #ceph
[10:04] * pentabular (~sean@adsl-70-231-141-17.dsl.snfc21.sbcglobal.net) has joined #ceph
[10:05] * nolan_ (~nolan@2001:470:1:41:20c:29ff:fe9a:60be) has joined #ceph
[10:05] * mdxi_ (~mdxi@74-95-29-182-Atlanta.hfc.comcastbusiness.net) has joined #ceph
[10:05] * laevar_ (~jochen@laevar.de) has joined #ceph
[10:05] * Meths_ (rift@2.25.193.120) has joined #ceph
[10:06] * dilemma (~dilemma@2607:fad0:32:a02:1e6f:65ff:feac:7f2a) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * ninkotech (~duplo@89.177.137.231) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * jbd_ (~jbd_@34322hpv162162.ikoula.com) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * dmick (~dmick@2607:f298:a:607:71b4:1e55:8139:f272) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * mdxi (~mdxi@74-95-29-182-Atlanta.hfc.comcastbusiness.net) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * sagewk (~sage@2607:f298:a:607:219:b9ff:fe40:55fe) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * yehudasa (~yehudasa@2607:f298:a:607:484d:925a:d3d:e9ca) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * nolan (~nolan@2001:470:1:41:20c:29ff:fe9a:60be) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * laevar (~jochen@laevar.de) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * Meths (rift@2.25.193.120) Quit (synthon.oftc.net graviton.oftc.net)
[10:06] * nolan_ is now known as nolan
[10:24] * gohko (~gohko@natter.interq.or.jp) has joined #ceph
[10:38] * loicd (~loic@178.20.50.225) has joined #ceph
[10:39] * MikeMcClurg (~mike@62.200.22.2) has joined #ceph
[10:40] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[10:48] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[10:49] * pentabular (~sean@adsl-70-231-141-17.dsl.snfc21.sbcglobal.net) has left #ceph
[11:33] * The_Bishop (~bishop@e179020222.adsl.alicedsl.de) has joined #ceph
[11:41] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[11:48] * guerby (~guerby@nc10d-ipv6.tetaneutral.net) has joined #ceph
[11:49] * KevinPerks (~Adium@38.122.20.226) Quit (Read error: Operation timed out)
[11:53] * tryggvil (~tryggvil@163-60-19-178.xdsl.simafelagid.is) Quit (Quit: tryggvil)
[12:02] * gaveen (~gaveen@112.135.148.69) Quit (Ping timeout: 480 seconds)
[12:04] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) has joined #ceph
[12:06] * The_Bishop (~bishop@e179020222.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[12:10] * gaveen (~gaveen@112.135.166.16) has joined #ceph
[12:17] * tryggvil (~tryggvil@rtr1.tolvusky.sip.is) has joined #ceph
[12:20] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Ping timeout: 480 seconds)
[12:22] * The_Bishop (~bishop@e179020222.adsl.alicedsl.de) has joined #ceph
[12:32] * tryggvil (~tryggvil@rtr1.tolvusky.sip.is) Quit (Quit: tryggvil)
[12:39] * yoshi (~yoshi@p37219-ipngn1701marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[12:51] * BManojlovic (~steki@87.110.183.173) Quit (Remote host closed the connection)
[12:51] * BManojlovic (~steki@87.110.183.173) has joined #ceph
[13:08] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[13:09] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[13:17] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[13:41] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[13:49] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[13:53] * bugfixer (~dchambers@42gis175.gulftel.com) has joined #ceph
[13:57] * deepsa (~deepsa@122.172.2.82) Quit (Ping timeout: 480 seconds)
[14:03] * mtk (~mtk@ool-44c35bb4.dyn.optonline.net) Quit (Remote host closed the connection)
[14:15] * BManojlovic (~steki@87.110.183.173) Quit (Ping timeout: 480 seconds)
[14:20] * mtk (~mtk@ool-44c35bb4.dyn.optonline.net) has joined #ceph
[14:34] * gaveen (~gaveen@112.135.166.16) Quit (Remote host closed the connection)
[14:42] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[14:50] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[15:00] * aliguori (~anthony@cpe-70-123-140-180.austin.res.rr.com) has joined #ceph
[15:01] * LarsFronius_ (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[15:01] * LarsFronius_ (~LarsFroni@testing78.jimdo-server.com) Quit (Remote host closed the connection)
[15:02] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) Quit (Read error: No route to host)
[15:02] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[15:06] * nhorman (~nhorman@hmsreliant.think-freely.org) has joined #ceph
[15:16] * deepsa (~deepsa@122.172.2.82) has joined #ceph
[15:26] * guerby (~guerby@nc10d-ipv6.tetaneutral.net) Quit (Ping timeout: 480 seconds)
[15:33] * sagelap (~sage@76.89.177.113) Quit (Ping timeout: 480 seconds)
[15:40] * sagelap (~sage@85.sub-70-197-146.myvzw.com) has joined #ceph
[15:42] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[15:50] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[15:57] * cblack101 (86868949@ircip4.mibbit.com) has joined #ceph
[15:59] * slang (~slang@2607:f298:a:607:a0b7:b301:3fb7:9a08) has joined #ceph
[16:06] * loicd (~loic@178.20.50.225) Quit (Ping timeout: 480 seconds)
[16:14] * tryggvil (~tryggvil@193.107.86.97) has joined #ceph
[16:17] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Quit: Leaving.)
[16:18] <elder> I have now discovered that the current master branch of ceph is not working for me. I get "Operation not permitted" errors when attempting to create rbd images.
[16:19] <elder> Switching to the stable branch doesn't have this problem. I believe this behavior began to happen sometime yesterday.
[16:19] <elder> Unfortunately I had to assume this was the fault of my own changes.
[16:31] * glowell (~Adium@68.170.71.123) has joined #ceph
[16:32] * loicd (~loic@2a01:e35:2eba:db10:120b:a9ff:feb7:cce0) has joined #ceph
[16:42] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[16:50] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[16:57] * Cube1 (~Adium@cpe-76-95-223-199.socal.res.rr.com) has joined #ceph
[17:02] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[17:06] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) Quit (Quit: Leseb)
[17:11] * allsystemsarego (~allsystem@188.27.164.159) has joined #ceph
[17:17] * LarsFronius_ (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[17:22] <elder> sagelap, do you know of any benefit of that le32_add_cpu() function? Just curious, I don't see the point, though maybe it just tidies up the code at the call sites.
[17:22] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) has joined #ceph
[17:23] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) Quit (Ping timeout: 480 seconds)
[17:23] * LarsFronius_ is now known as LarsFronius
[17:24] * glowell (~Adium@68.170.71.123) Quit (Quit: Leaving.)
[17:24] * KevinPerks (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[17:36] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[17:43] <joao> elder, looks like it's just aesthetics
[17:44] <joao> le32_add_cpu() seems to do pretty much the same as what was there before
[17:44] <elder> That's what I figured. I just wondered whether some architecture might have an instruction that might make that operation more efficient or something.
[17:45] <joao> yeah, maybe
[17:46] <elder> Well, maybe, but I don't see anybody overriding that definition, so I think it's aesthetics, like you said.
[17:46] <joao> elder, according to lxr, generic.h is the only one defining it
[17:46] <joao> yeah
[17:46] <joao> what you just said :p
[17:50] * sagelap (~sage@85.sub-70-197-146.myvzw.com) Quit (Ping timeout: 480 seconds)
[17:50] * Cube1 (~Adium@cpe-76-95-223-199.socal.res.rr.com) Quit (Quit: Leaving.)
[17:52] * KevinPerks (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[17:52] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) Quit (Read error: Connection reset by peer)
[17:52] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[17:53] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) Quit ()
[17:54] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[17:56] * tryggvil (~tryggvil@193.107.86.97) Quit (Ping timeout: 480 seconds)
[17:58] * mengesb (~bmenges@servepath-gw3.servepath.com) Quit (Quit: Leaving.)
[18:10] * tryggvil (~tryggvil@193.107.86.97) has joined #ceph
[18:10] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) Quit (Quit: Leaving.)
[18:11] * sagelap (~sage@103.sub-70-197-147.myvzw.com) has joined #ceph
[18:14] * danieagle (~Daniel@186.214.56.10) has joined #ceph
[18:15] <sagelap> elder, joshd: may be related to teh osd cap changes from yesterday?
[18:16] <sagelap> joao, elder: yeah, aesthetics, and harmless, and someone went to the effort of sending the patch
[18:17] <joao> indeed, that's a nice thing to see happening :)
[18:17] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) has joined #ceph
[18:19] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) has joined #ceph
[18:20] <elder> No complaint or criticism was intended, sagelap, I was just curious if there was a technical rather than aesthetic reason for those functions.
[18:20] <sagelap> yeah not that i know of :)
[18:21] <elder> I approve of the readability improvement.
[18:21] <elder> Or at least how it makes it look simpler...
[18:22] * tryggvil (~tryggvil@193.107.86.97) Quit (Ping timeout: 480 seconds)
[18:22] <joao> sagelap, do you see any reason why we should keep the gv quorum feature on the single-paxos monitor?
[18:25] * tryggvil (~tryggvil@193.107.86.97) has joined #ceph
[18:25] * gaveen (~gaveen@112.135.165.238) has joined #ceph
[18:26] * Cube (~Adium@12.248.40.138) has joined #ceph
[18:27] * glowell (~Adium@38.122.20.226) has joined #ceph
[18:30] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) Quit (Quit: Leaving.)
[18:32] <sagelap> which part? the _gv directories / mapping can go away, if that's what you mean
[18:32] * jlogan1 (~Thunderbi@72.5.59.176) has joined #ceph
[18:33] <sagelap> and the quorum thing where it waits for all paxos instances to recover becomes obsolete because there is only one, right?
[18:33] <joao> yeah
[18:36] <joao> sagelap, what is the accepted correct way to make an older monitor to be able to communicate with the newer?
[18:36] <joao> just changing the probe message compat version?
[18:36] <joao> header version, as well, I suppose
[18:37] <joao> or defining a quorum feature, much like we did for wip-mon-gv?
[18:37] <joao> both maybe?
[18:38] <joao> actually, going to change the probe message header version, given that I changed the message's payload format (removed the slurp stuff)
[18:47] <elder> Anybody know why I keep getting failures in the packaging of my ceph client builds?
[18:47] * chutzpah (~chutz@100.42.98.5) has joined #ceph
[18:47] <joao> ah, sagelap, nevermind
[18:48] <elder> dpkg-deb: building package `linux-tools-3.5.0-rc1-ceph-00037-gdf6ad1f' in `../linux-tools-3.5.0-rc1-ceph-00037-gdf6ad1f_3.5.0-rc1-ceph-00037-gdf6ad1f-1_amd64.deb'.
[18:48] <elder> warning: dpkg-gencontrol: package linux-image-3.5.0-rc1-ceph-00037-gdf6ad1f: unused substitution variable ${shlibs:Depends}
[18:48] <elder> dpkg-deb: building package `linux-image-3.5.0-rc1-ceph-00037-gdf6ad1f' in `../linux-image-3.5.0-rc1-ceph-00037-gdf6ad1f_3.5.0-rc1-ceph-00037-gdf6ad1f-1_amd64.deb'.
[18:48] <elder> dpkg-deb (subprocess): data member: internal gzip write error: 'Invalid argument'
[18:48] * yehudasa_ (~yehudasa@38.122.20.226) has joined #ceph
[18:48] <elder> error: dpkg-deb: subprocess from tar -cf returned error exit status 2
[18:48] <joao> I misread one of the gv patches; looks like the feature goes directly onto the messenger's policy
[18:51] <joao> elder, sorry, no idea :\
[18:51] <joao> don't think I have ever seen it, and google failed me
[18:53] * tryggvil (~tryggvil@193.107.86.97) Quit (Ping timeout: 480 seconds)
[18:54] * sagelap (~sage@103.sub-70-197-147.myvzw.com) Quit (Ping timeout: 480 seconds)
[18:54] * sagelap (~sage@38.122.20.226) has joined #ceph
[18:55] * KevinPerks (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[18:57] * sileht (~sileht@sileht.net) Quit (Quit: WeeChat 0.3.8)
[18:58] * joshd (~joshd@2607:f298:a:607:221:70ff:fe33:3fe3) has joined #ceph
[18:59] * sileht (~sileht@sileht.net) has joined #ceph
[19:00] * KevinPerks (~Adium@38.122.20.226) has joined #ceph
[19:02] <sagelap> wip-coverity-common branch is pushed, fi someone wants to skim it before i merge!
[19:04] <joao> sagelap, do we have any sort of criteria on addressing stuff from coverity, or just go through them and fix them is enough?
[19:06] * Guest8533 (~sean@adsl-70-231-141-17.dsl.snfc21.sbcglobal.net) has joined #ceph
[19:07] <sjust> sagelap: lookin
[19:18] * KevinPerks (~Adium@38.122.20.226) Quit (Ping timeout: 480 seconds)
[19:28] <sjust> sagelap: looks ok
[19:29] * gaveen (~gaveen@112.135.165.238) Quit (Remote host closed the connection)
[19:29] * maelfius (~mdrnstm@66.209.104.107) has joined #ceph
[19:31] * maelfius (~mdrnstm@66.209.104.107) Quit ()
[19:32] * tryggvil (~tryggvil@193.107.86.97) has joined #ceph
[19:57] * gaveen (~gaveen@112.135.165.238) has joined #ceph
[20:05] * The_Bishop_ (~bishop@e179021006.adsl.alicedsl.de) has joined #ceph
[20:12] * The_Bishop (~bishop@e179020222.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[20:14] * tryggvil_ (~tryggvil@193.107.86.97) has joined #ceph
[20:14] * tryggvil (~tryggvil@193.107.86.97) Quit (Read error: Connection reset by peer)
[20:14] * tryggvil_ is now known as tryggvil
[20:15] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) has joined #ceph
[20:16] * tryggvil_ (~tryggvil@193.107.86.97) has joined #ceph
[20:16] * tryggvil (~tryggvil@193.107.86.97) Quit (Read error: Connection reset by peer)
[20:16] * tryggvil_ is now known as tryggvil
[20:17] * tryggvil_ (~tryggvil@193.107.86.97) has joined #ceph
[20:17] * tryggvil (~tryggvil@193.107.86.97) Quit (Read error: Connection reset by peer)
[20:17] * tryggvil_ is now known as tryggvil
[20:20] * lofejndif (~lsqavnbok@28IAAHZFL.tor-irc.dnsbl.oftc.net) has joined #ceph
[20:21] * The_Bishop_ (~bishop@e179021006.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[20:22] * Meths_ is now known as Meths
[20:23] * loicd (~loic@2a01:e35:2eba:db10:120b:a9ff:feb7:cce0) Quit (Quit: Leaving.)
[20:25] * loicd (~loic@magenta.dachary.org) has joined #ceph
[20:25] * tryggvil (~tryggvil@193.107.86.97) Quit (Ping timeout: 480 seconds)
[20:32] * The_Bishop_ (~bishop@e179021006.adsl.alicedsl.de) has joined #ceph
[20:34] * bugfixer (~dchambers@42gis175.gulftel.com) Quit (Quit: Leaving)
[20:41] * LarsFronius (~LarsFroni@testing78.jimdo-server.com) Quit (Quit: LarsFronius)
[21:05] * jstrunk (~quassel@sophie.ma.utexas.edu) has joined #ceph
[21:05] * jmlowe (~Adium@c-71-201-31-207.hsd1.in.comcast.net) Quit (Quit: Leaving.)
[21:07] * Cube (~Adium@12.248.40.138) Quit (Ping timeout: 480 seconds)
[21:07] * MikeMcClurg (~mike@62.200.22.2) Quit (Ping timeout: 480 seconds)
[21:07] * Cube (~Adium@12.248.40.138) has joined #ceph
[21:08] <jstrunk> How does ceph calculate its capacity? I have 4 250GB OSDs and 4 1.1TB OSDs, but ceph status reports 8358 GB avail. I expected about 5500 GB at most.
[21:09] <joshd> it goes by the equivalent of 'df' on the osd data directories
[21:09] <joshd> and it's reporting raw capacity, since different pools can have different replication levels
[21:12] <jstrunk> joshd: sorry about that, I was misreading my disk sizes multiple times. My large OSDs have 1.8TB. It adds up. Thank you.
[21:12] <joshd> no problem
[21:13] * jstrunk (~quassel@sophie.ma.utexas.edu) Quit (Remote host closed the connection)
[21:19] * tryggvil (~tryggvil@rtr1.tolvusky.sip.is) has joined #ceph
[21:23] * joao (~JL@89.181.153.232) Quit (Ping timeout: 480 seconds)
[21:27] * gaveen (~gaveen@112.135.165.238) Quit (Remote host closed the connection)
[21:39] * BManojlovic (~steki@195.13.166.253) has joined #ceph
[21:44] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[21:44] * slang (~slang@2607:f298:a:607:a0b7:b301:3fb7:9a08) Quit (Quit: slang)
[21:46] * loicd (~loic@magenta.dachary.org) has joined #ceph
[21:48] * lofejndif (~lsqavnbok@28IAAHZFL.tor-irc.dnsbl.oftc.net) Quit (Ping timeout: 480 seconds)
[21:49] * jtang (~jtang@79.97.135.214) has joined #ceph
[21:50] * slang (~slang@38.122.20.226) has joined #ceph
[22:03] * slang (~slang@38.122.20.226) Quit (Ping timeout: 480 seconds)
[22:05] * MikeMcClurg (~mike@cpc10-cmbg15-2-0-cust205.5-4.cable.virginmedia.com) has joined #ceph
[22:06] * glowell (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[22:10] * elder (~elder@c-71-195-31-37.hsd1.mn.comcast.net) Quit (Read error: Connection reset by peer)
[22:13] * elder (~elder@c-71-195-31-37.hsd1.mn.comcast.net) has joined #ceph
[22:32] * nhorman (~nhorman@hmsreliant.think-freely.org) Quit (Quit: Leaving)
[22:34] * tryggvil (~tryggvil@rtr1.tolvusky.sip.is) Quit (Quit: tryggvil)
[22:35] * The_Bishop_ (~bishop@e179021006.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[22:46] * mtk (~mtk@ool-44c35bb4.dyn.optonline.net) Quit (Quit: Leaving)
[22:47] * The_Bishop_ (~bishop@e179021006.adsl.alicedsl.de) has joined #ceph
[22:51] <elder> joshd, does the user space code read /sys/bus/rbd/devices/N/size?
[22:51] <elder> joshd, I'm wondering if commit a51aa0c0 is a bug that needs to be ported back or not.
[22:53] <joshd> elder: no, I don't think so
[22:53] <elder> OK. What about updating the image size after a refresh?
[22:53] <elder> 93a24e0
[22:54] <joshd> if that affects the reported capacity, it should be backported
[22:55] <elder> It will.
[22:55] <elder> I'll add it tot he list.
[22:55] <elder> Well, wait a minute, where does the reported capacity come from?
[22:55] <elder> Nevermind, I'll just add it.
[22:56] * allsystemsarego (~allsystem@188.27.164.159) Quit (Quit: Leaving)
[22:56] <joshd> set_capacity is called somewhere like the old refresh_snap() iirc
[22:56] <elder> Yes.
[22:56] * tryggvil (~tryggvil@163-60-19-178.xdsl.simafelagid.is) has joined #ceph
[22:57] <elder> And that happens after a refresh. So (still) will add that patch to the list for back-porting.
[22:59] <joshd> sounds good
[23:02] * Guest8533 (~sean@adsl-70-231-141-17.dsl.snfc21.sbcglobal.net) Quit (Quit: Guest8533)
[23:05] <elder> joshd, another one. Is this important to the user space code? Sending the updated header object version number?
[23:05] <elder> a71b891b
[23:08] <joshd> no, it looks like that doesn't matter
[23:08] <elder> OK, thanks.
[23:08] * sagelap (~sage@38.122.20.226) Quit (Ping timeout: 480 seconds)
[23:13] * sagelap (~sage@2607:f298:a:607:4050:be7d:df57:b639) has joined #ceph
[23:25] * slang (~slang@216.3.101.62) has joined #ceph
[23:31] * John (~john@astound-69-42-3-170.ca.astound.net) has joined #ceph
[23:43] * jtang (~jtang@79.97.135.214) Quit (Read error: Connection reset by peer)
[23:44] <John> What are "journal aio" and journal dio" settings? What do they do?
[23:45] <sjust> journal aio enables using libaio for asyncronous writes to the journal
[23:45] <sjust> it should be more efficient
[23:45] <sjust> but it does require aio compiled in
[23:45] <sjust> it also implies journal dio
[23:45] <sjust> journal dio enables directio to the journal
[23:46] <sjust> much more efficient than not
[23:46] <sjust> but can't be used in some cases
[23:46] <sjust> dio is on by default, aio is off by default, but aio is usually a good idea
[23:46] <John> Is aio compiled in with stable releases?
[23:46] <sjust> yeah,
[23:46] <sjust> *yeah
[23:47] <cblack101> Question: On my journal drive (SSD) should I use noop/deadline/cfq for the scheduler, or does it matter?
[23:48] <sjust> journal block align requires writes to align to block boundaries, can cost some space efficiency, but is faster (implied by directio)
[23:48] <sjust> cblack101: I don't think we've done much analysis in that area, my guess is that it probably does matter :)
[23:48] <John> Can aio and dio be on at the same time, or is it one or the other? What is implied if dio is false?
[23:49] <sjust> aio implies/requires dio, if dio is false, we just do writes with periodic fsyncs
[23:49] <cblack101> In which case, I will test all of them, thanks sjut
[23:49] <sjust> cblack101: k, let us know how it goes!
[23:49] * slang (~slang@216.3.101.62) Quit (Ping timeout: 480 seconds)
[23:49] <cblack101> *sjust
[23:50] <John> Can you give a brief overview of "journal block align"?
[23:51] <sjust> it's really just a detail for directio, directio (and therefore aio) require rights to be block aligned
[23:51] <sjust> *writes
[23:52] <John> So if dio is on, block align must be on too?
[23:52] <sjust> yeah
[23:53] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) has joined #ceph
[23:54] * Leseb (~Leseb@bea13-1-82-228-104-16.fbx.proxad.net) Quit ()
[23:55] <John> And "journal max write bytes" ?
[23:55] <sjust> that's the limit on how large a chunk the journal will write at a time
[23:56] <sjust> adjusting that probably won't help for the most part
[23:56] <sjust> similarly, journal_max_write_entries limits the number of entries the journal can write at a time
[23:56] <sjust> increasing that might help in some cases
[23:57] <sjust> journal queue max * limit the journal op and size throttles (limits on number of ops and total size of ops allowed to be in the queue)
[23:57] <John> How about "journal queue max ops"?
[23:58] <sjust> journal queue max ops limits the journal op throttle, that is, limits the number of ops allowed in the queue
[23:58] <sjust> similarly with bytes
[23:58] * slang (~slang@216.3.101.62) has joined #ceph

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