Switch to DuckDuckGo Search
   March 30, 2020  
< | 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
[00:20:35] *** KeiraT- <KeiraT-!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[00:21:23] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[00:23:31] *** KeiraT- is now known as KeiraT
[00:28:03] <Mokou> https://code.illumos.org/c/illumos-gate/+/464 change pushed
[00:28:17] <Mokou> should I do something to send a review request to the mailing list?
[00:34:50] <LeftWing> Yes, I would do that
[00:35:02] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Read error: Connection timed out)
[00:35:29] <LeftWing> Perhaps include the contents of the comment you put on the change, too
[00:35:34] <LeftWing> It includes a lot of useful context!
[00:37:58] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[00:40:23] <Mokou> roger :D
[00:45:38] *** shuryanc[m] <shuryanc[m]!shuryancma@gateway/shell/matrix.org/x-zjykksvpweynlmxr> has quit IRC (Ping timeout: 246 seconds)
[00:46:37] *** shuryanc[m] <shuryanc[m]!shuryancma@gateway/shell/matrix.org/x-xsukfdfqnuglvqws> has joined #illumos
[00:50:02] *** joltman <joltman!znc@gateway/vpn/privateinternetaccess/joltman> has quit IRC (Ping timeout: 256 seconds)
[00:51:01] *** elegast <elegast!~elegast@83-161-172-44.mobile.xs4all.nl> has quit IRC (Remote host closed the connection)
[00:54:30] *** joltman <joltman!~znc@unaffiliated/joltman> has joined #illumos
[01:20:35] <sjorge> rmustacc: is there some mdb magic I can do to check if the driver enables all the offloading bits? I might chuck it in one of my boxes to see if it still works. Driver man lage doesn’t mention anything about offloading. (FreeBSD and Linux Driver manpages do thoigh)
[01:26:05] <rmustacc> sjorge: Which driver?
[01:26:31] <rmustacc> We should make that a bit clearer in dladm (and probably allow that to be toggled).
[02:43:29] *** alanc <alanc!~alanc@inet-hqmc01-o.oracle.com> has quit IRC (Remote host closed the connection)
[02:43:57] *** alanc <alanc!~alanc@inet-hqmc01-o.oracle.com> has joined #illumos
[02:53:41] *** wl_ <wl_!~wl_@2605:6000:1b0c:46ad::87c> has quit IRC (Remote host closed the connection)
[03:14:35] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Quit: Leaving)
[03:35:10] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[03:41:37] *** Tempt <Tempt!~avenger@unaffiliated/tempt> has quit IRC (Ping timeout: 264 seconds)
[04:20:40] *** Tempt <Tempt!~avenger@unaffiliated/tempt> has joined #illumos
[04:20:40] *** ChanServ sets mode: +o Tempt
[05:32:51] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[05:33:02] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[05:36:08] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Client Quit)
[05:59:39] *** arnoldoree <arnoldoree!~arnoldore@113.210.199.98> has quit IRC (Ping timeout: 256 seconds)
[06:20:15] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Quit: Leaving)
[06:38:23] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[06:39:02] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Remote host closed the connection)
[06:53:57] *** neuroserve <neuroserve!~toens@ip-178-202-217-1.hsi09.unitymediagroup.de> has quit IRC (Ping timeout: 265 seconds)
[06:54:01] *** cartwright <cartwright!~chatting@gateway/tor-sasl/cantstanya> has quit IRC (Write error: Connection reset by peer)
[06:54:01] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Write error: Connection reset by peer)
[06:54:14] *** neuroserve <neuroserve!~toens@ip-178-202-217-1.hsi09.unitymediagroup.de> has joined #illumos
[06:56:53] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[06:58:09] *** cartwright <cartwright!~chatting@gateway/tor-sasl/cantstanya> has joined #illumos
[07:33:13] *** neuroserve <neuroserve!~toens@ip-178-202-217-1.hsi09.unitymediagroup.de> has quit IRC (Ping timeout: 264 seconds)
[07:41:06] *** Asgaroth_ <Asgaroth_!~Asgaroth@93.107.144.95> has quit IRC (Read error: Connection reset by peer)
[07:42:08] *** Asgaroth_ <Asgaroth_!~Asgaroth@109.76.10.56> has joined #illumos
[08:13:14] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 240 seconds)
[08:33:46] *** neuroserve <neuroserve!~toens@ip-94-114-252-237.unity-media.net> has joined #illumos
[08:34:25] *** pwinder <pwinder!~pwinder@86.11.191.8> has joined #illumos
[08:37:18] *** BOKALDO <BOKALDO!~BOKALDO@81.198.158.89> has joined #illumos
[08:41:58] <pwinder> /msg NickServ identify J@2minec
[09:34:27] *** jimklimov <jimklimov!~jimklimov@78.80.224.132> has joined #illumos
[09:45:11] <sjorge> rmustacc: bnxe according to hcl
[09:46:33] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[09:52:01] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a80:b324:97ed:5468:4fef:c249> has joined #illumos
[10:05:33] <LeftWing> pwinder: You may want to change that password...
[10:12:07] *** Asgaroth_ <Asgaroth_!~Asgaroth@109.76.10.56> has quit IRC (Read error: Connection reset by peer)
[10:12:32] *** Asgaroth_ <Asgaroth_!~Asgaroth@109.76.10.56> has joined #illumos
[10:16:49] *** man_u <man_u!~manu@89-92-19-81.hfc.dyn.abo.bbox.fr> has joined #illumos
[10:33:08] *** psarria <psarria!~phyre___@78.30.17.247> has joined #illumos
[10:46:55] *** andy_js <andy_js!~andy@51.146.99.40> has joined #illumos
[11:20:54] *** cartwright <cartwright!~chatting@gateway/tor-sasl/cantstanya> has quit IRC (Remote host closed the connection)
[11:25:55] *** cartwright <cartwright!~chatting@gateway/tor-sasl/cantstanya> has joined #illumos
[11:35:52] <pwinder> :-)
[11:40:00] *** pwinder99 <pwinder99!560bbf08@86.11.191.8> has joined #illumos
[11:43:14] *** pwinder <pwinder!~pwinder@86.11.191.8> has left #illumos ("Leaving")
[11:43:24] *** pwinder99 <pwinder99!560bbf08@86.11.191.8> has left #illumos
[11:43:53] *** pwinder99 <pwinder99!560bbf08@86.11.191.8> has joined #illumos
[11:44:03] *** pwinder99 <pwinder99!560bbf08@86.11.191.8> has quit IRC (Remote host closed the connection)
[11:45:32] *** pwinder <pwinder!560bbf08@86.11.191.8> has joined #illumos
[12:24:41] *** nbhauke <nbhauke!~hauke@55d45eef.access.ecotel.net> has joined #illumos
[13:03:41] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a80:b324:97ed:5468:4fef:c249> has quit IRC (Ping timeout: 252 seconds)
[13:06:10] *** tsoome_ <tsoome_!~tsoome@91.209.240.229> has joined #illumos
[13:10:58] *** tsoome_ <tsoome_!~tsoome@91.209.240.229> has quit IRC (Client Quit)
[13:14:08] *** Guest56538 <Guest56538!~void@ip4d16b7c2.dynamic.kabel-deutschland.de> has quit IRC (Ping timeout: 256 seconds)
[13:14:09] *** yomi <yomi!~void@ip4d16bd50.dynamic.kabel-deutschland.de> has joined #illumos
[13:14:32] *** yomi is now known as Guest40707
[13:17:42] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a80:b324:97ed:5468:4fef:c249> has joined #illumos
[13:21:35] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[13:31:11] *** Guest40707 <Guest40707!~void@ip4d16bd50.dynamic.kabel-deutschland.de> has quit IRC (Ping timeout: 252 seconds)
[13:40:58] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has joined #illumos
[13:46:26] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has quit IRC (Ping timeout: 256 seconds)
[13:58:18] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has joined #illumos
[14:23:36] *** psarria <psarria!~phyre___@78.30.17.247> has quit IRC (Remote host closed the connection)
[14:48:35] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a80:b324:97ed:5468:4fef:c249> has quit IRC (Remote host closed the connection)
[14:52:41] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a80:b324:97ed:5468:4fef:c249> has joined #illumos
[14:59:54] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[15:14:54] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has quit IRC (Quit: How to defend yourself against an attacker armed with a mathematician)
[15:15:37] *** denk_ <denk_!~denis@devel.tambov.ru> has joined #illumos
[15:15:48] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 240 seconds)
[15:15:48] *** glasspelican <glasspelican!~quassel@2607:5300:201:3100::325> has quit IRC (Ping timeout: 240 seconds)
[15:15:48] *** jordan[m]11 <jordan[m]11!~jordantim@2607:3f00:11:24::3> has quit IRC (Ping timeout: 240 seconds)
[15:15:48] *** wilbury <wilbury!otis@netbsd/developer/otis> has quit IRC (Ping timeout: 240 seconds)
[15:15:49] *** leah2 <leah2!~leah@vuxu.org> has quit IRC (Ping timeout: 240 seconds)
[15:15:49] *** mgerdts <mgerdts!~textual@2600:6c44:c7f:ec89:c54c:48c2:f7f4:f79a> has quit IRC (Ping timeout: 240 seconds)
[15:15:49] *** denk <denk!~denis@devel.tambov.ru> has quit IRC (Ping timeout: 240 seconds)
[15:15:49] *** minus <minus!~minus@2001:41d0:e:35b::1> has quit IRC (Ping timeout: 240 seconds)
[15:16:32] *** wilbury <wilbury!otis@m68k.sinclairql.eu> has joined #illumos
[15:16:34] *** jordan[m]11 <jordan[m]11!~jordantim@104.166.94.34> has joined #illumos
[15:17:38] *** minus <minus!~minus@ns379866.ip-5-196-70.eu> has joined #illumos
[15:17:41] *** glasspelican <glasspelican!~quassel@2607:5300:201:3100::325> has joined #illumos
[15:17:44] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[15:18:14] *** denk_ is now known as denk
[15:18:34] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has joined #illumos
[15:18:36] *** leah2 <leah2!~leah@vuxu.org> has joined #illumos
[16:25:56] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[16:37:44] *** shuryanc1 <shuryanc1!shuryancsi@gateway/shell/matrix.org/x-evjbinlwlwvsbntv> has left #illumos ("Kicked by @appservice-irc:matrix.org : Idle for 30+ days")
[16:50:50] *** teutat3s <teutat3s!jhotmatrix@gateway/shell/matrix.org/x-hlwqonrajdwdpcek> has left #illumos ("Kicked by @appservice-irc:matrix.org : Idle for 30+ days")
[16:51:08] <EisNerd> quick question howto import a zpool only temporary
[16:51:32] <EisNerd> so that it isn't automatically imported on next boot?
[16:54:49] <EisNerd> another maybe stupid question, shouldn't a zpool find all its devices automatically no matter how the devices are actually named? At least this is how understood it
[16:57:44] <denk> to avoid importing pool next boot do not save it in the system cache file (what zpool import does by default) or specify alternative cachefile
[16:57:53] <denk> zpool import -o cachefile=none <pool>
[17:00:16] <ptribble> or zpool import -t
[17:03:09] <EisNerd> thx
[17:08:35] *** mpana <mpana!~mpana@193.226.149.49> has quit IRC (Ping timeout: 260 seconds)
[17:08:56] *** Exagone313 <Exagone313!exa@3131313.xyz> has quit IRC (Quit: see ya!)
[17:09:14] *** FUZxxl <FUZxxl!~fuz@fuz.su> has quit IRC (Ping timeout: 240 seconds)
[17:09:50] *** jduggan <jduggan!~thom@s.tankengine.org.uk> has quit IRC (Ping timeout: 246 seconds)
[17:09:56] *** Exagone313 <Exagone313!exa@3131313.xyz> has joined #illumos
[17:11:55] *** edef <edef!edef@NixOS/user/edef> has quit IRC (Ping timeout: 252 seconds)
[17:11:56] *** danmcd <danmcd!~danmcd@static-71-174-113-16.bstnma.fios.verizon.net> has joined #illumos
[17:11:58] *** shuryanc[m] <shuryanc[m]!shuryancma@gateway/shell/matrix.org/x-xsukfdfqnuglvqws> has quit IRC (Ping timeout: 240 seconds)
[17:12:37] *** FUZxxl <FUZxxl!~fuz@fuz.su> has joined #illumos
[17:13:01] *** FUZxxl is now known as Guest43706
[17:13:39] *** mpana <mpana!~mpana@193.226.149.49> has joined #illumos
[17:14:43] *** shuryanc[m] <shuryanc[m]!shuryancma@gateway/shell/matrix.org/x-bywjmbpeozqkxqmh> has joined #illumos
[17:16:30] <EisNerd> hm the smb server is still unexpected slow, according to bonnie my pool can do 500MB/s and my direct network link 1.5GB/s, so the limit should be the source with its 1GBit link, but I only get 60-70MB/s, I would expect 110
[17:19:25] *** Guest43706 <Guest43706!~fuz@fuz.su> has quit IRC (Read error: Connection reset by peer)
[17:26:34] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Ping timeout: 240 seconds)
[17:31:19] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has joined #illumos
[17:33:37] *** jduggan <jduggan!~thom@s.tankengine.org.uk> has joined #illumos
[17:34:44] *** FUZxxl <FUZxxl!~fuz@fuz.su> has joined #illumos
[17:35:08] *** FUZxxl is now known as Guest88890
[17:35:58] *** edef__ <edef__!~edef@NixOS/user/edef> has joined #illumos
[17:35:58] *** edef__ is now known as edef
[17:40:19] *** dsockwell <dsockwell!~dsockwell@mnemonic.hightechlow.life> has quit IRC (Ping timeout: 260 seconds)
[17:40:31] *** neirac <neirac!~neirac@pc-215-101-46-190.cm.vtr.net> has joined #illumos
[17:41:19] *** dsockwell <dsockwell!~dsockwell@163.172.219.97> has joined #illumos
[17:45:08] *** scoobybrrrrjesus <scoobybrrrrjesus!sid271506@gateway/web/irccloud.com/x-okfqkyxyskekyfnw> has quit IRC (*.net *.split)
[17:45:08] *** GrahamPerrin[m] <GrahamPerrin[m]!grahamperr@gateway/shell/matrix.org/x-xnqahvxcanonltzy> has quit IRC (*.net *.split)
[17:45:08] *** funnyguy <funnyguy!funnyguy@2600:3c02::f03c:91ff:fe56:7f51> has quit IRC (*.net *.split)
[17:45:08] *** Aram <Aram!~aram@9front.army> has quit IRC (*.net *.split)
[17:45:08] *** vaxsquid <vaxsquid!~vaxsquid@banana.fish> has quit IRC (*.net *.split)
[17:45:08] *** CustosLimen <CustosLimen!sid53872@gateway/web/irccloud.com/x-tuvbzcmgrlykxmbb> has quit IRC (*.net *.split)
[17:45:33] *** scoobybrrrrjesus <scoobybrrrrjesus!sid271506@gateway/web/irccloud.com/x-okfqkyxyskekyfnw> has joined #illumos
[17:45:33] *** GrahamPerrin[m] <GrahamPerrin[m]!grahamperr@gateway/shell/matrix.org/x-xnqahvxcanonltzy> has joined #illumos
[17:45:33] *** CustosLimen <CustosLimen!sid53872@gateway/web/irccloud.com/x-tuvbzcmgrlykxmbb> has joined #illumos
[17:45:33] *** funnyguy <funnyguy!funnyguy@2600:3c02::f03c:91ff:fe56:7f51> has joined #illumos
[17:45:33] *** Aram <Aram!~aram@9front.army> has joined #illumos
[17:45:33] *** vaxsquid <vaxsquid!~vaxsquid@banana.fish> has joined #illumos
[17:46:17] *** CustosLimen <CustosLimen!sid53872@gateway/web/irccloud.com/x-tuvbzcmgrlykxmbb> has quit IRC (Max SendQ exceeded)
[17:48:13] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has quit IRC (Ping timeout: 264 seconds)
[17:48:35] *** neuroserve <neuroserve!~toens@ip-94-114-252-237.unity-media.net> has quit IRC (Ping timeout: 252 seconds)
[17:48:42] *** shuryanc[m] <shuryanc[m]!shuryancma@gateway/shell/matrix.org/x-bywjmbpeozqkxqmh> has quit IRC (Ping timeout: 260 seconds)
[17:48:46] *** Guest88890 is now known as FUZxxl
[17:48:49] *** GrahamPerrin[m] <GrahamPerrin[m]!grahamperr@gateway/shell/matrix.org/x-xnqahvxcanonltzy> has quit IRC (Ping timeout: 240 seconds)
[17:48:58] *** neuroserve <neuroserve!~toens@ip-178-202-217-1.hsi09.unitymediagroup.de> has joined #illumos
[17:49:14] *** morgib[m] <morgib[m]!morgibmatr@gateway/shell/matrix.org/x-oicjvqigjjgounnh> has quit IRC (Ping timeout: 256 seconds)
[17:49:18] *** psydruid <psydruid!psydruidma@gateway/shell/matrix.org/x-htevfuzoyiexxiqx> has quit IRC (Ping timeout: 260 seconds)
[17:49:19] *** Ericson2314 <Ericson2314!ericson231@gateway/shell/matrix.org/x-bnhipyxpdtwadpwu> has quit IRC (Ping timeout: 260 seconds)
[17:49:21] *** TelegramBridge[m <TelegramBridge[m!telegramt2@gateway/shell/matrix.org/x-seizosokamowrdws> has quit IRC (Ping timeout: 272 seconds)
[17:49:50] *** CustosLimen <CustosLimen!sid53872@gateway/web/irccloud.com/x-vcwuzipazzodmxgn> has joined #illumos
[17:51:06] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has joined #illumos
[17:53:27] *** vaxsquid <vaxsquid!~vaxsquid@banana.fish> has quit IRC (Ping timeout: 240 seconds)
[17:53:37] *** vaxsquid <vaxsquid!~vaxsquid@banana.fish> has joined #illumos
[17:57:47] *** neirac <neirac!~neirac@pc-215-101-46-190.cm.vtr.net> has quit IRC (Remote host closed the connection)
[18:01:50] *** skaldix <skaldix!~meow@67.216.58.158> has joined #illumos
[18:02:35] *** neirac <neirac!~neirac@pc-215-101-46-190.cm.vtr.net> has joined #illumos
[18:05:38] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has joined #illumos
[18:24:07] <sjorge> Aah ldap the gift that keeps on giving... looks like I will need some idmap magic to store SID’s too for smbsrv
[18:24:16] <sjorge> There te some hits in the idmap manpage
[18:24:38] <sjorge> But nothing concrete 🤔
[18:25:37] <sjorge> EisNerd: I think practically you should zee around 90MB/s on a 1Gbit link over smb
[18:25:59] <sjorge> At least I have never seen more with other products like EMC Isilon or samba either
[18:41:46] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:1a80:b324:97ed:5468:4fef:c249> has quit IRC (Ping timeout: 240 seconds)
[18:46:19] *** shuryanc[m] <shuryanc[m]!shuryancma@gateway/shell/matrix.org/x-lpjjgequcvqetiby> has joined #illumos
[18:53:32] *** nbhauke <nbhauke!~hauke@55d45eef.access.ecotel.net> has quit IRC (Quit: nbhauke)
[19:10:36] *** pwinder <pwinder!560bbf08@86.11.191.8> has quit IRC (Ping timeout: 240 seconds)
[19:18:05] <richlowe> sjorge: so if you figure it out, fix the manual page :)
[19:19:11] *** GrahamPerrin[m] <GrahamPerrin[m]!grahamperr@gateway/shell/matrix.org/x-jdkxrhmunmxotpsp> has joined #illumos
[19:19:11] *** Ericson2314 <Ericson2314!ericson231@gateway/shell/matrix.org/x-nmrjbvxptvbswokq> has joined #illumos
[19:19:11] *** TelegramBridge[m <TelegramBridge[m!telegramt2@gateway/shell/matrix.org/x-odzrddbhuyvgyacx> has joined #illumos
[19:19:11] *** morgib[m] <morgib[m]!morgibmatr@gateway/shell/matrix.org/x-pixmtjlalqwnruhb> has joined #illumos
[19:19:11] *** psydruid <psydruid!psydruidma@gateway/shell/matrix.org/x-fdvyzphfyucfchjv> has joined #illumos
[19:20:51] *** nbhauke <nbhauke!~hauke@55d45eef.access.ecotel.net> has joined #illumos
[19:28:44] *** andy_js <andy_js!~andy@51.146.99.40> has quit IRC (Read error: Connection reset by peer)
[19:29:23] *** andy_js <andy_js!~andy@51.146.99.40> has joined #illumos
[19:36:40] *** khng300 <khng300!~khng300@unaffiliated/khng300> has quit IRC (Quit: ZNC 1.7.5 - https://znc.in)
[19:38:00] *** khng300 <khng300!~khng300@unaffiliated/khng300> has joined #illumos
[20:13:54] *** deejam- is now known as deejam
[20:29:55] <bahamat> Is there a name for fmri path components reported by fmtopo? Like "motherboard=0/chip=0/core=0". Devices? Nodes?
[20:30:20] *** jimklimov <jimklimov!~jimklimov@78.80.224.132> has quit IRC (Quit: Leaving.)
[20:31:56] <bahamat> Using xml output I see FRU and resource, so maybe resource is the word I'm looking for.
[20:36:32] <rmustacc> No, it's not. That's a specific term.
[20:36:54] <bahamat> Is there something better?
[20:36:54] <rmustacc> At least, in terms of topo.
[20:37:40] <rmustacc> I don't think there's a generic name used by human. We have a name for it in the library.
[20:37:47] <rmustacc> What's the higher-level thing?
[20:38:11] <rmustacc> Are you trying to figure out how to communicate something in parituclar or looking for a generic name for everything?
[20:38:22] <bahamat> I'm writing a cmon plugin to report temps and I'm parsing the fmri to convert them to prometheus labels.
[20:38:46] <bahamat> So I want the variable names I'm using to be meaningful.
[20:39:38] <rmustacc> So ignore information that you have there for a second, if you were administering the system, what would you want to know?
[20:39:53] <rmustacc> Presumably which socket the sensor relates to and which subpart?
[20:40:04] <bahamat> Yeah.
[20:40:06] <rmustacc> (Keep in mind different chips have different schemes for sensors)
[20:40:23] <rmustacc> So I think the question then is do you want that in terms of things that the OS knows or the things that you the operator cares about.
[20:40:45] <rmustacc> For example, do you want a label that matches the silkscreen on the motherboard or do you want a label that matches what the OS calls that CPU socket?
[20:41:25] <bahamat> More what the operator cares about, but I'm also biasing for additional information in case it might be useful.
[20:43:31] <rmustacc> So, the topo FMRI describes something that topo knows about there (though that path isn't necessairily stable). So I'd probably look a the other properties that you have at different levels and figure out what to call it. For example, you could name it based on the chip and core, but keep in mind that your platforms don't have a good way to match that back to the smbios label on the silkscreen today.
[20:43:34] <bahamat> Considering that the silk screen can be arbitrary, I tend to not rely on it much.
[20:44:37] <bahamat> So do you mean that just because it says "chip=0", that doesn't actually mean that it's in socket 0?
[20:45:41] <rmustacc> It is what the OS considers socket 0, but that has no relation to a silk screen.
[20:46:06] <rmustacc> silk screens usually are started numbering at 1, but Intel/AMD start numbering at zero.
[20:46:47] <bahamat> OK, I get what you mean.
[20:47:00] <bahamat> I'm less concerned about it being off by one than I am that they're in the proper order.
[20:47:32] <rmustacc> I never felt comfortable guaranteeing that on platforms.
[20:47:33] <bahamat> like if 0 is silk screened as 1 and 1 is silk screened as 2, then I'm ok with emitting 0 and 1
[20:47:50] <bahamat> But if 0 is ss 1 and 1 is ss 0 that would be a problem.
[20:48:13] <rmustacc> The OS can only be as consistent as the BIOS/fw vendor is.
[20:48:55] <bahamat> Are the fmtopo index numbers in the same order as smbios index numbers?
[20:49:02] <rmustacc> They are orthogonal.
[20:49:11] <rmustacc> Topo is derived from the apic IDs.
[20:49:30] <rmustacc> So what the CPU actually tells us about topoogy.
[20:49:33] <rmustacc> *topology
[20:51:36] <bahamat> OK, ultimately I want to be able to potentially tell someone to replace a part based on this information.
[20:52:17] <rmustacc> Well, if something is overtemp the host will force power off.
[20:52:45] <rmustacc> But yes, trying to match up the two is an unsovled problem in x86 and ultimately is a per-vendor thing that relies on your vendor getting the fw right.
[20:54:59] <bahamat> In that case we'd probably end up using the sel to identify the part, and we'd be able to match that up with the temperature history, regardless of the index numbers
[20:55:06] <bahamat> I think overall there's two scenarios
[20:55:33] <bahamat> 1. There's a cpu part malfunctioning and therefore running hot, in which case we want to replace that cpu
[20:56:26] <bahamat> 2. The environment is hot, in which case all cpus should be running within some standard deviation of each other
[20:56:45] <rmustacc> I'm not sure I really follow 1 exactly. If 1 is a problem, the sysem will shut down, but 2 makes sense.
[20:57:04] <rmustacc> But I guess if there was a mfg defect with the heatsink or something else.
[20:57:13] <bahamat> But we might be able to detect a problem cpu running elevated before it reaches the threshhold.
[20:58:07] <bahamat> But yeah, there can be something affecting just one cpu, or something affecting all cpus.
[20:58:57] <rmustacc> Tracking the data definitely makes sense. That's why I added those sensors.
[20:59:16] <bahamat> I'm not even 100% sure how I might use this yet. But one thing I'm interested in is if the whole DC is growing warmer.
[21:00:17] <bahamat> I also want to factor this in to part failure rates.
[21:00:40] <bahamat> especially for disks
[21:03:00] *** nde <nde!uid414739@gateway/web/irccloud.com/x-miqslkqsynxspyip> has joined #illumos
[21:06:25] <bahamat> As an aside, I have a system with usb attached disks that fmtopo reports there is sensors for, but gets an error retrieving the props. If there's any data I can provide to fix that at some point, I'd be happy to.
[21:08:54] <bahamat> So since I can't really correlate the fmtopo layout to smbios, I think just using the fmtopo identifiers is fine, which brings me back to my original question of what to call the path segments :-)
[21:09:36] <rmustacc> I wouldn't name it based on that. I would translate it into something that makes sense for the component.
[21:09:48] <rmustacc> And as your aside, there's not enough info to really advise.
[21:11:01] <neirac> LeftWing I sort of done adding flock to golang syscall for illumos, I ended up separating Solaris and Illumos syscall definitions in src/syscall I think is better that way
[21:11:28] <bahamat> Well for disks, the most meaningful element is the serial number.
[21:11:47] <rmustacc> Right, I'd do whatever made sense on a per-component basis. You do have the CPU serial number on some platforms too.
[21:12:01] <rmustacc> I wouldn't really encode the FMRI except as some of the information in the FMRI is useful.
[21:12:14] <rmustacc> (But then it should be accessible as properties from the topo snapshot)
[21:12:23] <bahamat> Oh? Is that available on amd? I don't see it on any of the intel systems I have.
[21:12:41] <bahamat> Or I'm not looking in the right place.
[21:13:11] <rmustacc> Are you looking at a Xeon?
[21:13:22] <rmustacc> The Intel PPIN support was something I added a while back.
[21:13:30] <rmustacc> AMD doesn't have it yet (and it's only available on Rome)
[21:13:45] <bahamat> Yeah
[21:14:12] <rmustacc> Unless something broke it, should show up in topo, but probably on the socket, not on the lower levels.
[21:14:51] <sjorge> richlowe side form, idmap can also pull the info form ldapclient there is not much of a hint it should probably read and write to it though? *shrug*
[21:14:55] <sjorge> Lets get freeradius working first I guess
[21:15:23] <bahamat> And sometimes I don't get any sensor results from fmtopo
[21:16:16] <rmustacc> I assume you're running as root?
[21:16:19] <bahamat> OH, I forgot to wildcard it
[21:16:29] <bahamat> I did 'sensor=temp' not '*sensor=temp'
[21:20:17] <rmustacc> Ah, yeah. I've done that quite often.
[21:20:29] <bahamat> For example, I've got this: https://gist.github.com/bahamat/cf26712b09aed815e7d21b469e3148a4
[21:20:36] <bahamat> but the serial number isn't in that.
[21:22:12] <rmustacc> bahamat: It's not in the sensor node.
[21:22:39] <rmustacc> But also, an 1101 is a Sandy Bridge part right?
[21:22:56] <bahamat> I don't see it when I take off the query string either.
[21:23:39] <bahamat> Like doing just `hc://.../motherboard=0/chip=1` doesn't show a serial for cpus.
[21:23:59] <rmustacc> This is a Sandy Bridge part right?
[21:24:15] <bahamat> Intel(r) Xeon(r) CPU E5-2670 0 @ 2.60GHz
[21:24:32] <rmustacc> So, yes. That's before Intel introduced it.
[21:24:33] <bahamat> I don't have the code names memorized :-)
[21:24:47] <bahamat> Ok, so I should see it if I look in something later.
[21:24:54] <bahamat> Let me check one of the JBs.
[21:25:34] <rmustacc> The BIOS can lock that down, so it's not guaranteed either.
[21:25:59] <rmustacc> So using that or the logical OS chip / core ID.
[21:26:43] <bahamat> OK, yeah, serial is there on the newer ones
[21:26:55] <bahamat> Great, that's super helpful.
[21:26:58] <rmustacc> It is a somewhat synthetic value,b ut at least it's something.
[21:32:23] *** jenelizabeth <jenelizabeth!~jenelizab@185.92.25.103> has quit IRC (Ping timeout: 256 seconds)
[21:32:57] *** superjen96 <superjen96!~jenelizab@85.203.46.129> has joined #illumos
[21:34:36] <bahamat> Yeah. I mean, if we detect an error by whatever means, we'll be counting on the OS to tell us enough info about it to tell dcops what needs replacing. Where possible we'll cross reference that with the SEL.
[21:37:08] <bahamat> If there's a colon, slash, equal sign, or question mark in a serial number, what does fmtopo do? Does it URI encode it?
[21:39:42] <rmustacc> In the topo libraries they're just C 'char *'s.
[21:42:15] <bahamat> Ok, that's something I'll need to address at some point then.
[21:42:31] <rmustacc> I don't know what interface you're using, so that's something to consider.
[21:44:28] <bahamat> I'm doing `fmtopo '*?sensor=temp'` to get the list of temp sensors available, then `fmtopo -P facility.reading :fmri | awk ...` to get the value.
[21:45:02] <bahamat> But I'm also trying to parse the fmri to get some meaningful prometheus labels.
[21:45:39] <rmustacc> I would not do that.
[21:45:49] <rmustacc> I would write a proper topo consumer and use all the semantic information and properties.
[21:46:12] <rmustacc> What you are doing will be rather brittle and will be more confused hwen you're doing it on AMD.
[21:46:44] <rmustacc> Being a topo consumer means you can get access to the properties and other aspects (though note all of this is unstable).
[21:47:17] <rmustacc> I think a solution would be much more robust if you were actually reading data from parent nodes that had properties rather than awking your way to victory.
[21:47:34] <LeftWing> I reckon it'd be pretty easy to knock together a Rust binding to libtopo
[21:51:06] <bahamat> I would definitely like to do that, but that's going to take me longer.
[21:51:22] <bahamat> I'm not even sure how to properly compile rust for the global zone yet.
[21:52:27] *** man_u <man_u!~manu@89-92-19-81.hfc.dyn.abo.bbox.fr> has quit IRC (Quit: man_u)
[21:53:10] <bahamat> Using rust form pkgsrc, it binaries link against pkgsrc's libgcc
[21:54:04] <rmustacc> In your case, just ship it and edit the link paths?
[21:54:09] <rmustacc> run paths
[21:56:08] <bahamat> Yeah, that's one option.
[21:57:07] <bahamat> It seems to me that eventually it'd be useful to be able to have rust in gate.
[21:58:31] <rmustacc> Sure, but that's kind of orthogonal.
[22:00:16] <LeftWing> Yeah I would collect and ship the `libgcc_s.so` alongside the binary and `elfedit` in an `$ORIGIN/../lib` link path
[22:00:25] <LeftWing> Like we did for sdc-node
[22:01:03] <LeftWing> Depending on which crates you use, Rust can be in general much less sensitive than, say, a C program
[22:01:08] <LeftWing> to which machine it's built on
[22:01:27] <bahamat> Right.
[22:01:46] <LeftWing> Obviously if you use the OpenSSL crate and don't use the static linking mode, that will present the usual challenges haha
[22:02:00] <bahamat> And especially if you're using no_std
[22:02:14] <LeftWing> Well if you're using no_std it will link against very little
[22:02:23] <bahamat> yeah, ssl's going to be a nightmare.
[22:02:41] <LeftWing> Eh, if you use the static linking mode of the OpenSSL crate it'll just fold the current OpenSSL into the binary
[22:02:48] <LeftWing> That's what they do with the rustup binaries
[22:02:51] <LeftWing> Seems to work OK
[22:07:48] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: Leaving)
[22:07:58] *** tsoome <tsoome!~tsoome@5726-dd23-deed-f4c1-2f80-4a40-07d0-2001.sta.estpak.ee> has joined #illumos
[22:09:21] <bahamat> We do have libcrypto.a and libssl.a in pkgsrc, so yeah, that probably the best way.
[22:34:05] *** BOKALDO <BOKALDO!~BOKALDO@81.198.158.89> has quit IRC (Quit: Leaving)
[22:49:10] *** nbhauke <nbhauke!~hauke@55d45eef.access.ecotel.net> has quit IRC (Quit: nbhauke)
[22:57:11] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 260 seconds)
[22:58:45] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[23:03:28] *** rui <rui!~rui@a89-154-24-173.cpe.netcabo.pt> has joined #illumos
[23:13:45] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[23:29:44] *** andy_js <andy_js!~andy@51.146.99.40> has quit IRC (Quit: andy_js)
[23:31:57] *** morgib[m] <morgib[m]!morgibmatr@gateway/shell/matrix.org/x-pixmtjlalqwnruhb> has left #illumos ("Kicked by @appservice-irc:matrix.org : Idle for 30+ days")
[23:52:41] *** rui <rui!~rui@a89-154-24-173.cpe.netcabo.pt> has quit IRC (Quit: rui)
top

   March 30, 2020  
< | 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 | >