Switch to DuckDuckGo Search
   July 6, 2019
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31

Toggle Join/Part | bottom
[00:37:06] <gitomat> [illumos-gate] 11256 prstat man page says -v and -m are different, but they aren't -- Brian Bennett <brian.bennett at joyent dot com>
[00:48:41] *** andy_js <andy_js!~andy@94.12.186.242> has quit IRC (Quit: andy_js)
[01:13:04] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[01:33:17] *** DaQatz <DaQatz!~DB@2601:187:8400:5::83c> has quit IRC (Remote host closed the connection)
[01:36:43] *** Qatz <Qatz!~DB@2601:187:8400:5::83c> has joined #illumos
[01:50:09] *** cantstanya <cantstanya!~chatting@gateway/tor-sasl/cantstanya> has quit IRC (Remote host closed the connection)
[01:51:58] *** cantstanya <cantstanya!~chatting@gateway/tor-sasl/cantstanya> has joined #illumos
[01:54:52] *** _Tenchi_ <_Tenchi_!~phil@d-207-255-80-203.paw.cpe.atlanticbb.net> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[02:39:39] *** basvdlei <basvdlei!~basvdlei@2001:980:a4c3:1:80f2:2bd8:393d:9cf5> has quit IRC (Ping timeout: 250 seconds)
[02:42:09] *** khng300 <khng300!~khng300@unaffiliated/khng300> has quit IRC (Ping timeout: 248 seconds)
[03:05:39] *** khng300 <khng300!~khng300@unaffiliated/khng300> has joined #illumos
[03:20:01] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[03:20:10] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[03:21:21] *** baojg <baojg!~baojg@162.243.44.213> has joined #illumos
[03:24:38] *** baojg <baojg!~baojg@162.243.44.213> has quit IRC (Remote host closed the connection)
[03:25:12] *** baojg <baojg!~baojg@162.243.44.213> has joined #illumos
[04:33:26] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has joined #illumos
[05:00:50] <gitomat> [illumos-gate] 11020 want SMB2 ioctl FSCTL_SRV_COPYCHUNK -- Gordon Ross <gwr at nexenta dot com>
[05:00:51] <gitomat> [illumos-gate] 11021 more SMB2 ioctls -- Gordon Ross <gwr at nexenta dot com>
[05:00:52] <gitomat> [illumos-gate] 11022 Offload Data Transfer (ODX) support -- Gordon Ross <gwr at nexenta dot com>
[05:50:58] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has quit IRC (Remote host closed the connection)
[06:45:27] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Ping timeout: 245 seconds)
[06:49:32] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has quit IRC (Remote host closed the connection)
[07:04:04] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[07:20:21] *** BOKALDO <BOKALDO!~BOKALDO@87.110.147.49> has joined #illumos
[07:22:55] *** cumulus_ <cumulus_!Elite22250@gateway/shell/elitebnc/x-onqbfotbwqgxycid> has quit IRC (Ping timeout: 258 seconds)
[08:13:53] *** basvdlei <basvdlei!~basvdlei@lamuella.xs4all.nl> has joined #illumos
[08:38:15] *** alanc <alanc!~alanc@129.157.69.51> has quit IRC (Remote host closed the connection)
[08:38:43] *** alanc <alanc!~alanc@129.157.69.51> has joined #illumos
[09:06:36] *** kev009 <kev009!~kev009@ip72-222-200-117.ph.ph.cox.net> has quit IRC (Remote host closed the connection)
[09:09:49] *** kev009 <kev009!~kev009@ip72-222-200-117.ph.ph.cox.net> has joined #illumos
[10:13:21] *** baojg_ <baojg_!~baojg@162.243.44.213> has joined #illumos
[10:13:22] *** baojg <baojg!~baojg@162.243.44.213> has quit IRC (Ping timeout: 245 seconds)
[10:21:47] *** Asgaroth <Asgaroth!~Asgaroth@212.2.172.228> has joined #illumos
[10:28:00] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 258 seconds)
[11:08:55] *** trisk <trisk!~trisk@2601:196:4700:3f0b:8:20ff:fe3e:890e> has quit IRC (Remote host closed the connection)
[11:14:24] *** trisk <trisk!~trisk@2601:196:4700:3f0b:8:20ff:fe3e:890e> has joined #illumos
[11:20:06] *** andy_js <andy_js!~andy@94.12.186.242> has joined #illumos
[11:20:55] *** baojg <baojg!~baojg@162.243.44.213> has joined #illumos
[11:21:43] *** baojg_ <baojg_!~baojg@162.243.44.213> has quit IRC (Ping timeout: 245 seconds)
[11:41:14] *** baojg <baojg!~baojg@162.243.44.213> has quit IRC (Remote host closed the connection)
[12:15:15] *** AndrewG <AndrewG!~andrew@a20.cucumber.me.uk> has joined #illumos
[12:16:52] *** AndrewG <AndrewG!~andrew@a20.cucumber.me.uk> has quit IRC (Client Quit)
[12:20:01] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[12:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[12:29:20] *** AndrewG <AndrewG!~andrew@a20.cucumber.me.uk> has joined #illumos
[12:49:37] *** CVLTCMK0 <CVLTCMK0!~Ident@ool-2f1257d8.dyn.optonline.net> has quit IRC (Ping timeout: 248 seconds)
[12:55:40] *** CVLTCMK0 <CVLTCMK0!~Ident@ool-2f1257d8.dyn.optonline.net> has joined #illumos
[13:06:36] *** w8hd <w8hd!~kimc@cpe-65-25-122-137.neo.res.rr.com> has joined #illumos
[13:17:11] *** phyre__ <phyre__!~phyre___@31.211.191.164> has joined #illumos
[13:24:22] *** shruti <shruti!shruti@nat/redhat/x-coubdpmngywheaqi> has quit IRC (Ping timeout: 246 seconds)
[13:24:37] *** shruti <shruti!shruti@nat/redhat/x-uypdlklbfvneqxwx> has joined #illumos
[13:37:41] *** merzo <merzo!~merzo@244-35-133-95.pool.ukrtel.net> has joined #illumos
[13:49:01] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has joined #illumos
[13:53:28] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has quit IRC (Ping timeout: 258 seconds)
[14:55:15] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has joined #illumos
[14:59:47] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has quit IRC (Ping timeout: 258 seconds)
[15:06:09] *** phyre__ <phyre__!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 248 seconds)
[15:24:45] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has joined #illumos
[15:25:23] *** shruti <shruti!shruti@nat/redhat/x-uypdlklbfvneqxwx> has quit IRC (Ping timeout: 244 seconds)
[15:29:18] *** Kurlon <Kurlon!~Kurlon@h69-130-65-1.nansme.dsl.dynamic.tds.net> has quit IRC (Ping timeout: 258 seconds)
[15:30:54] *** shruti <shruti!shruti@nat/redhat/x-pznpojgvhxkcybdq> has joined #illumos
[16:03:48] *** shruti <shruti!shruti@nat/redhat/x-pznpojgvhxkcybdq> has quit IRC (Ping timeout: 245 seconds)
[16:04:04] *** shruti <shruti!shruti@nat/redhat/x-psxyltssenaiztrq> has joined #illumos
[16:18:21] <gitomat> [illumos-gate] 11293 cfgadm: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:21:17] <gitomat> [illumos-gate] 11294 abi: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:24:22] <gitomat> [illumos-gate] 11295 lp: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:26:29] <gitomat> [illumos-gate] 11311 devctl: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:29:14] <gitomat> [illumos-gate] 11312 dladm: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:32:00] *** Asgaroth <Asgaroth!~Asgaroth@212.2.172.228> has quit IRC (Ping timeout: 272 seconds)
[16:32:41] <gitomat> [illumos-gate] 11318 svc: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:36:34] <gitomat> [illumos-gate] 11320 diskinfo: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:37:27] *** AndrewG <AndrewG!~andrew@a20.cucumber.me.uk> has quit IRC (Quit: Leaving.)
[16:40:09] <gitomat> [illumos-gate] 11321 eeprom: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[16:40:50] *** AndrewG <AndrewG!~andrew@a20.cucumber.me.uk> has joined #illumos
[16:44:12] <igork> kmdb ::findleaks -v problems: https://cl.ly/0e0efcb5b492
[16:44:50] <igork> i have hw server with 256GB RAM, configured nmi to kmdb prompt because have no dump device
[16:45:14] <igork> produce nmi and have kmdb prompt, but ::findleaks -v - failed
[16:45:21] <igork> how to try to fix it?
[16:46:00] <igork> how to add more space to buffer by /etc/system, if it is possible
[16:46:03] <rmustacc> Don't use kmdb
[16:46:17] <rmustacc> It really isn't designed for the memory needed for findleaks.
[16:46:19] <igork> i can't produce full dump
[16:46:40] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has joined #illumos
[16:46:42] <rmustacc> You have less than 256 GiB of storage?
[16:46:45] <igork> i have very strange issues with LACP
[16:46:48] <igork> yes
[16:46:53] <igork> i have no dump device
[16:47:04] <igork> 120GB SSD for boot only
[16:47:06] <rmustacc> I'd suggest getting one or using another approach
[16:47:20] <rmustacc> Unfortunately, kmdb really isn't designed to have a lot of memory, which findleaks needs
[16:47:42] <igork> ok, could you please suggest something?
[16:47:51] <igork> how to try to debug leakes?
[16:47:53] <tsoome> default dump would only save kernel pages, so you would not need full 256GB:)
[16:48:09] <rmustacc> Right.
[16:48:29] <rmustacc> General advice is to get a crash dump off the system. Then run findleaks with mdb.
[16:48:53] <igork> ok
[16:48:54] <rmustacc> If you suspect it's a small thing happening a lot of the time, you don't need user pages and the kernel pages shouldn't grow that high.
[16:48:56] <tsoome> can we still dump to NFS swap?
[16:49:12] <rmustacc> I don't think we've wired up the networking stack for dump operation.
[16:49:22] <rmustacc> Though one could look at adding polling support for those.
[16:49:26] <igork> tsoome: if you have no network - you have no nfs
[16:49:46] <tsoome> well, in that case it is time to get some disk drive.
[16:49:51] <rmustacc> I'm just trying to suggest what exists with the capabilities that we have today.
[16:49:59] <igork> rmustacc: thanks
[16:51:01] <igork> and in this case only one way possible - add dump device and produce core dump to it and check findleaks - right?
[16:51:24] <rmustacc> No, that's not the only way possible.
[16:51:33] <igork> what another one?
[16:51:35] <rmustacc> But it's certainly the fastest and most reliable.
[16:51:44] <rmustacc> Well, I mean, you manually look at allocations and frees.
[16:51:55] <rmustacc> Try to minimize the operation that you think is leaking memory and walk caches, etc.
[16:52:02] <rmustacc> There's no single algorithm.
[16:52:08] <rmustacc> findleaks is much more time effective.
[16:52:11] <igork> i need try to collect data for ixgbe + mac layer - all kernel space
[16:52:39] <rmustacc> Start with kmastat and see where the memory is.
[16:52:47] <igork> strange issue with LACP huawai +
[16:52:55] <rmustacc> Is there GiB of memory in streams dblk caches?
[16:53:08] <rmustacc> I haven't met a buggy lacp switch yet.
[16:53:15] <rmustacc> Erm, one that isn't full of bugs.
[16:53:22] <rmustacc> And we do a good job of triggering some since we're not Linux.
[16:53:39] <igork> rmustacc: do yo haev experience with huawai switches + LACP on illumos side?
[16:53:43] <rmustacc> But none of those have been memory leaks.
[16:53:58] <rmustacc> Never used a huawei switch.
[16:54:20] <rmustacc> Today's politics probably means I won't for some time to come.
[16:54:26] <igork> let me try explain issue. i have 4 links in aggr on illumos side and 2 huawai in MLAG
[16:54:34] <rmustacc> Nope, don't want to know about it today, sorry.
[16:54:44] <igork> ok, thanks for update
[16:54:46] <rmustacc> Look at the LACP state.
[16:54:52] <rmustacc> Read the spec.
[16:55:04] <igork> LACP on illumos side is fine
[16:55:55] <igork> but of LACP drop on huawei side - it has side affect network stack on illumos - it kill ixgbe driver
[16:56:53] <tsoome> that is bad, driver should be able to recover.
[16:57:04] <igork> for example: i have no connections to onboard ixgbe interfaces, but in lofs i can see unregister/register for driver
[16:57:17] <igork> after illumos with LACP on huawei side
[16:57:27] <igork> s/illumos/issues
[16:58:09] <igork> yes - driver recovery only on one version over SFP+ what loaded in RAM and responded only to ping
[16:58:22] <igork> server is in stuck mode - no ssh connection to it
[16:58:30] <igork> only hard reset
[16:59:10] <igork> i need some ideas how to collect data and try identify the way - what is killing ixgbe + network stask
[16:59:13] <igork> *stack
[16:59:24] <igork> it is remote system
[17:00:00] <igork> i tried to ask to add hdd fro dump device - but it can't be done yet
[17:02:03] <tsoome> rmustacc: btw, did you see https://www.illumos.org/issues/11332 ?
[17:02:07] <igork> i have LACP dump traffic where point to issue on huawei side - it has broken LACP with timeout, but on server side SFP+ 10G aggr not working correctly after trash by LACP from huawei switch
[17:03:39] *** _Tenchi_ <_Tenchi_!~phil@d-207-255-80-203.paw.cpe.atlanticbb.net> has joined #illumos
[17:04:03] <tsoome> well, in such case I’d start with trying to see why it is failing to recover.
[17:04:18] <igork> it is what i try to do
[17:04:34] <igork> but server responded only to ping
[17:04:42] <igork> no serial console too
[17:04:53] <igork> no prompt for login
[17:04:56] <igork> nothing
[17:06:07] <tsoome> in such case you could at least create alternate network access.
[17:06:47] <igork> it is remote one where 4 links are in one aggr
[17:07:36] <tsoome> move one link out of aggregation or get network adapter.
[17:08:19] <tsoome> if it is having issues with 4-way aggregation, it hopefully should have the same issues with 2-way too.
[17:08:23] <igork> i asked about it too - not customer couldn't to do it
[17:08:28] *** _Tenchi_ <_Tenchi_!~phil@d-207-255-80-203.paw.cpe.atlanticbb.net> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[17:09:06] <tsoome> well, in that case, just get decent switches.
[17:09:18] <igork> in this way - you have remote black box with configured env ...
[17:09:59] <tsoome> of course it is about if the customer can afford to wait for fix.
[17:10:10] <igork> well, thanks for ideas - i'll try to find the way how to produce dump by nmi
[17:11:55] <igork> another problem - i have to wait a time when server will not respond
[17:12:44] <tsoome> it is random defect?
[17:12:47] <igork> LACP is broken after 3-5 days with high loads by networking writes/reads
[17:13:08] <igork> yes - random - in every 3-4-5-7 days
[17:13:21] <tsoome> even more fun...
[17:13:30] <igork> depend on indound traffic
[17:13:38] <igork> *inbound
[17:14:54] <igork> customer has a ticket to huawei and we are collecting logs from our side where we have confirmed issue on switches side
[17:18:51] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-8.net.upcbroadband.cz> has joined #illumos
[17:45:07] *** phyre__ <phyre__!~phyre___@31.211.191.164> has joined #illumos
[18:03:55] *** jocthbr <jocthbr!~salci@138-122-44-143.host.cicloti.com.br> has quit IRC (Quit: No Ping reply in 180 seconds.)
[18:05:14] *** jocthbr <jocthbr!~salci@138-122-44-143.host.cicloti.com.br> has joined #illumos
[18:06:05] *** phyre__ <phyre__!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 258 seconds)
[18:15:33] <gitomat> [illumos-gate] 11314 Provide psm_get_ipivect even on single CPU systems -- Gordon Ross <gwr at nexenta dot com>
[18:23:45] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Quit: 410 Gone)
[18:28:46] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #illumos
[18:30:17] *** phyre__ <phyre__!~phyre___@31.211.191.164> has joined #illumos
[18:36:51] *** kohju <kohju!~kohju@gw.justplayer.com> has quit IRC (Excess Flood)
[18:37:26] *** kohju <kohju!~kohju@gw.justplayer.com> has joined #illumos
[18:44:38] *** kujira <kujira!~null@71-36-99-73.ptld.qwest.net> has joined #illumos
[18:48:47] *** phyre__ <phyre__!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 245 seconds)
[19:49:55] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[20:14:52] *** squidgy <squidgy!~sqiodgy@97.99.89.242> has quit IRC (Quit: Leaving...)
[20:21:38] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[20:29:40] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[20:31:43] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[21:00:59] *** khng300 <khng300!~khng300@unaffiliated/khng300> has quit IRC (Quit: ZNC 1.7.3 - https://znc.in)
[21:02:10] *** khng300 <khng300!~khng300@unaffiliated/khng300> has joined #illumos
[21:11:39] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[21:34:15] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[21:46:32] *** BOKALDO <BOKALDO!~BOKALDO@87.110.147.49> has quit IRC (Quit: Leaving)
[22:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[22:20:09] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[22:36:10] *** ZOP <ZOP!~ZOP@phobos.wgops.com> has quit IRC (Ping timeout: 272 seconds)
[22:37:29] *** ZOP <ZOP!~ZOP@phobos.wgops.com> has joined #illumos
[22:41:57] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[22:50:26] *** gvm <gvm!~gvm@cpc91192-cmbg18-2-0-cust82.5-4.cable.virginm.net> has joined #illumos
[23:04:39] <gitomat> [illumos-gate] 11296 cmd-inet: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[23:06:28] <gitomat> [illumos-gate] 11297 chmod: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[23:14:38] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[23:27:07] *** jperkin <jperkin!~jperkin@custard.perkin.org.uk> has quit IRC (Ping timeout: 245 seconds)
[23:29:05] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 248 seconds)
[23:31:08] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[23:35:09] <gitomat> [illumos-gate] 11298 consadm: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[23:58:22] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has quit IRC (Ping timeout: 245 seconds)
top

   July 6, 2019
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31