Switch to DuckDuckGo Search
   May 2, 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
[00:07:31] *** lgtaube <lgtaube!~lgt@45.86.203.25> has quit IRC (*.net *.split)
[00:07:31] *** alanc <alanc!~alanc@inet-hqmc01-o.oracle.com> has quit IRC (*.net *.split)
[00:07:31] *** kovert <kovert!~kovert@204.141.173.249> has quit IRC (*.net *.split)
[00:07:31] *** ngchk1_ <ngchk1_!~ngchk1@b2b-92-50-91-166.unitymedia.biz> has quit IRC (*.net *.split)
[00:07:31] *** dlg <dlg!~dlg@toy.eait.uq.edu.au> has quit IRC (*.net *.split)
[00:07:31] *** jemershaw <jemershaw!~jemershaw@c-68-83-252-28.hsd1.pa.comcast.net> has quit IRC (*.net *.split)
[00:07:31] *** neuroserve <neuroserve!~toens@ip-88-152-243-25.hsi03.unitymediagroup.de> has quit IRC (*.net *.split)
[00:07:31] *** Tempt <Tempt!~avenger@unaffiliated/tempt> has quit IRC (*.net *.split)
[00:07:31] *** eki <eki!~eki@dsl-hkibng41-54f858-46.dhcp.inet.fi> has quit IRC (*.net *.split)
[00:07:31] *** lystra <lystra!~lystra@d53-64-11-169.nap.wideopenwest.com> has quit IRC (*.net *.split)
[00:07:31] *** BrownBear <BrownBear!~BrownBear@107.175.247.247> has quit IRC (*.net *.split)
[00:07:31] *** LeftWing <LeftWing!~jclulow@vortex.sysmgr.org> has quit IRC (*.net *.split)
[00:07:31] *** sensille <sensille!~arne@dsnsi.rzone.de> has quit IRC (*.net *.split)
[00:07:31] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has quit IRC (*.net *.split)
[00:07:33] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (*.net *.split)
[00:07:33] *** hemi770 <hemi770!~hemi666@unaffiliated/hemi770> has quit IRC (*.net *.split)
[00:07:33] *** glasspelican <glasspelican!~quassel@2607:5300:201:3100::325> has quit IRC (*.net *.split)
[00:07:33] *** dopplergange <dopplergange!~dop@192.111.134.213> has quit IRC (*.net *.split)
[00:07:33] *** vila <vila!~vila@laubervilliers-659-1-133-94.w80-15.abo.wanadoo.fr> has quit IRC (*.net *.split)
[00:07:33] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has quit IRC (*.net *.split)
[00:07:33] *** qyliss <qyliss!~qyliss@NixOS/user/qyliss> has quit IRC (*.net *.split)
[00:07:33] *** tomww <tomww!~tom@unaffiliated/tomww> has quit IRC (*.net *.split)
[00:07:33] *** Hazelesque <Hazelesque!~hazel@lopsa/member/hazelesque> has quit IRC (*.net *.split)
[00:07:33] *** tru_tru <tru_tru!~tru@157.99.90.140> has quit IRC (*.net *.split)
[00:07:33] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has quit IRC (*.net *.split)
[00:07:33] *** Tsesarevich <Tsesarevich!Tsesarevic@fluxbuntu/founder/joejaxx> has quit IRC (*.net *.split)
[00:07:33] *** perlgod <perlgod!~cullum@puffy.c0ffee.net> has quit IRC (*.net *.split)
[00:07:33] *** Cthulhux <Cthulhux!cthulhu@piratenpartei/ni/tux> has quit IRC (*.net *.split)
[00:07:43] *** Hazelesque <Hazelesque!~hazel@lopsa/member/hazelesque> has joined #illumos
[00:07:55] *** sensille <sensille!~arne@dsnsi.rzone.de> has joined #illumos
[00:08:03] *** tomww <tomww!~tom@gate.wagner-net.com> has joined #illumos
[00:08:10] *** LeftWing <LeftWing!~jclulow@vortex.sysmgr.org> has joined #illumos
[00:08:14] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has joined #illumos
[00:08:14] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[00:08:16] *** hemi770 <hemi770!~hemi666@46.23.78.25> has joined #illumos
[00:08:22] *** Cthulhux <Cthulhux!~cthulhu@rosaelefanten.org> has joined #illumos
[00:08:33] *** perlgod <perlgod!~cullum@puffy.c0ffee.net> has joined #illumos
[00:08:34] *** Tsesarevich <Tsesarevich!Tsesarevic@fluxbuntu/founder/joejaxx> has joined #illumos
[00:08:40] *** lgtaube <lgtaube!~lgt@45.86.203.25> has joined #illumos
[00:08:45] *** kovert <kovert!~kovert@204.141.173.249> has joined #illumos
[00:08:48] *** jemershaw <jemershaw!~jemershaw@c-68-83-252-28.hsd1.pa.comcast.net> has joined #illumos
[00:09:10] *** neuroserve <neuroserve!~toens@ip-88-152-243-25.hsi03.unitymediagroup.de> has joined #illumos
[00:10:32] *** BrownBear <BrownBear!~BrownBear@107.175.247.247> has joined #illumos
[00:10:53] *** dopplergange <dopplergange!~dop@192.111.134.213> has joined #illumos
[00:10:55] *** qyliss <qyliss!~qyliss@NixOS/user/qyliss> has joined #illumos
[00:11:11] *** glasspelican <glasspelican!~quassel@2607:5300:201:3100::325> has joined #illumos
[00:11:16] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has quit IRC (Ping timeout: 246 seconds)
[00:12:38] *** eki <eki!~eki@dsl-hkibng41-54f858-46.dhcp.inet.fi> has joined #illumos
[00:12:39] *** dlg <dlg!~dlg@toy.eait.uq.edu.au> has joined #illumos
[00:12:41] *** Tempt <Tempt!~avenger@unaffiliated/tempt> has joined #illumos
[00:12:44] *** ChanServ sets mode: +o Tempt
[00:13:01] *** tru_tru <tru_tru!~tru@157.99.90.140> has joined #illumos
[00:13:06] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has joined #illumos
[00:43:51] <pmooney> we should be patch-free to build tokio on the illumos (and solaris) target
[00:44:28] <pmooney> thanks to LeftWing for his persistent letter-writing campaigns to get us the maintainer access required
[00:44:41] <LeftWing> haha
[00:44:52] <LeftWing> I can write comments on GitHub for days
[00:44:58] <LeftWing> A core skill
[00:45:10] <pmooney> it only involved adopting three crates
[00:45:21] <pmooney> two and a half, by some measures
[00:45:23] <LeftWing> The Rust community has been very friendly, it must be said, and I am enjoying participating
[00:45:28] <pmooney> indeed!
[00:48:47] <rmustacc> Nice work!
[00:54:34] <jbk> is it still using epoll?
[00:54:56] <jbk> I know papertigers had some bugs w/ his event port implementation he never had a chance to hunt down
[00:54:59] <jbk> (tokio)
[00:57:07] *** toastersonerson <toastersonerson!~toastmatr@2a02:168:a407::4e7> has quit IRC (Ping timeout: 240 seconds)
[00:59:43] *** toastersonerson <toastersonerson!~toastmatr@85-195-235-249.fiber7.init7.net> has joined #illumos
[01:08:40] *** andy_js <andy_js!~andy@90.218.209.121> has quit IRC (Quit: andy_js)
[01:10:34] <LeftWing> I imagine there's still some epoll in there
[01:10:54] <LeftWing> In the limit I imagine we'll get event ports square away in mio/tokio/etc
[01:11:24] <LeftWing> I've been tinkering with adding PORT_SOURCE_FILE to notify-rs
[01:14:52] <gitomat> [illumos-gate] 12683 w(1) and whodo(1) should silently skip dirs it cannot open -- Jason King <jason.king at joyent dot com>
[01:19:03] <pmooney> jbk: yes, epoll for now
[01:46:55] *** src_builder <src_builder!~src_build@183.107.74.137.lunanode-rdns.com> has joined #illumos
[01:54:00] *** hemi770 <hemi770!~hemi666@unaffiliated/hemi770> has quit IRC (Quit: The life of a repo man is intense)
[01:54:33] *** hemi770 <hemi770!~hemi666@unaffiliated/hemi770> has joined #illumos
[02:17:51] *** neirac <neirac!~neirac@pc-4-149-45-190.cm.vtr.net> has joined #illumos
[02:21:21] *** src_builder <src_builder!~src_build@183.107.74.137.lunanode-rdns.com> has quit IRC (Quit: leaving)
[03:18:07] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Ping timeout: 240 seconds)
[03:21:07] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[03:40:40] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[03:47:24] *** bdha1 is now known as bdha
[04:50:52] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has quit IRC (Ping timeout: 258 seconds)
[05:22:31] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has joined #illumos
[05:28:10] *** arnoldoree <arnoldoree!~arnoldore@113.210.185.109> has joined #illumos
[06:18:00] *** BOKALDO <BOKALDO!~BOKALDO@91.105.124.91> has joined #illumos
[07:26:09] *** KungFuJesus <KungFuJesus!~adam@72.49.160.91> has joined #illumos
[07:26:13] <KungFuJesus> tsoome_: uh oh: /kernel/drv/sparcv9/mpt_sas: undefined symbol 'sata_split_model'
[07:26:36] <KungFuJesus> got that booting from the ISO, not sure this is going to go well, hah. Using the mpt_sas controller for sure
[07:27:34] <KungFuJesus> https://pastebin.com/YaXAY7fk
[07:28:27] <KungFuJesus> Also since when did Oracle/Sun attach the DVD ROM drive over the USB bus? According to the front bezel it's a "sata dvd-rom"
[07:28:46] <tsoome> KungFuJesus, can not tell anything about the iso itself
[07:29:25] <tsoome> that data dependency issue has been seen before, probably iso was created before the fix
[07:29:44] <KungFuJesus> aww man :(. Do you maybe have a newer ISO I can try?
[07:30:00] <tsoome> it seems to be working anyhow?
[07:30:25] <KungFuJesus> well, maybe, it's been sitting on that last step forever
[07:30:44] <KungFuJesus> and from what I remember, it sits there for hours before failing unless I boot from a flash drive instead
[07:31:01] <tsoome> hm, mkay.
[07:32:03] <tsoome> I use ldom myself...
[07:42:27] <KungFuJesus> there it goes: https://pastebin.com/QExN4APV
[07:43:27] <KungFuJesus> https://pastebin.com/MFmLPmCZ
[07:43:38] <KungFuJesus> I'm not having a ton of confidence in this ISO, ha
[07:45:55] <tsoome> Preparing text install image for use is about mounting lofi images
[07:48:43] <tsoome> it may be bad media/read errors, however, we would hope to see io errors reported
[07:48:48] <KungFuJesus> do you have a good way to convert this to a USB install like Oracle provides? I suspect the optical drive on this thing may have issues
[07:52:13] <tsoome> create VTOC on it, with ufs on s0, copy the iso over with .files and installboot -F ufs ..
[07:52:43] <tsoome> it may need something more but that might do for starters
[07:52:54] *** nbhauke <nbhauke!~hauke@55d475e1.access.ecotel.net> has joined #illumos
[07:53:48] <KungFuJesus> ah, I may need some more specific instructions for that. I haven't manually created a solaris partition table ever (or at least that I can remember)
[07:54:37] <tsoome> format -e, select disk, partition
[07:57:23] <KungFuJesus> ok, did partition, spat out a warning
[07:57:38] <igork> KungFuJesus: https://dilos-dev.atlassian.net/wiki/spaces/DS/pages/30048260/Prepare+boot+disk
[07:58:00] <KungFuJesus> https://pastebin.com/Q4BJ7NNn
[08:00:39] <KungFuJesus> both label and partition give me that warning
[08:00:39] <KungFuJesus> and drop me back to the same prompt
[08:00:39] <KungFuJesus> hmm, maybe I should just blow away the existing contents
[08:02:50] *** KungFuJe1us <KungFuJe1us!~adam@72.49.160.91> has joined #illumos
[08:03:32] *** KungFuJe1us <KungFuJe1us!~adam@72.49.160.91> has quit IRC (Client Quit)
[08:03:34] *** bacterio <bacterio!~bacterio@fsf/member/bacterio> has quit IRC (Ping timeout: 260 seconds)
[08:04:45] *** KungFuJe1us <KungFuJe1us!~adam@72.49.160.91> has joined #illumos
[08:04:55] <KungFuJe1us> sorry, connection dropped for a brief moment there
[08:06:17] *** KungFuJesus <KungFuJesus!~adam@72.49.160.91> has quit IRC (Ping timeout: 260 seconds)
[08:06:37] <KungFuJe1us> yeah it's still not letting me label the thing, is there something I'm missing here?
[08:06:42] *** nbhauke <nbhauke!~hauke@55d475e1.access.ecotel.net> has quit IRC (Quit: nbhauke)
[08:09:19] *** bacterio <bacterio!~bacterio@fsf/member/bacterio> has joined #illumos
[08:12:19] <KungFuJe1us> apparently just proceding to fdisk makes it do it anyway
[08:14:05] <KungFuJe1us> so what are the other steps?
[08:15:02] <tsoome> fdisk? sparc does not use fdisk.
[08:16:47] <KungFuJe1us> no from the format prompt
[08:17:00] <KungFuJe1us> I'm using x86 illumos to make the bootable image for SPARC
[08:30:21] <tsoome> nono, format -> disk will run fdisk and you can not have MBR on stick, sparc does not recognize it.
[08:33:54] <KungFuJe1us> I gotta get to bed but I should be able to resume this in the morning
[08:42:48] *** lystra <lystra!~lystra@d53-64-11-169.nap.wideopenwest.com> has joined #illumos
[08:54:48] *** arnoldoree <arnoldoree!~arnoldore@113.210.185.109> has quit IRC (Ping timeout: 272 seconds)
[08:57:55] *** arnoldoree <arnoldoree!~arnoldore@113.210.185.109> has joined #illumos
[09:01:14] *** fgudin <fgudin!fgudin@sdf-eu.org> has quit IRC (Ping timeout: 240 seconds)
[09:14:55] *** vgusev_ <vgusev_!~vgusev@188.187.60.230> has joined #illumos
[09:18:14] *** vgusev <vgusev!~vgusev@188.187.60.230> has quit IRC (Ping timeout: 260 seconds)
[09:30:03] *** ypankov <ypankov!~ypankov@178.34.26.10> has quit IRC (Quit: leaving)
[09:33:19] *** arnoldoree <arnoldoree!~arnoldore@113.210.185.109> has quit IRC (Ping timeout: 260 seconds)
[09:35:34] *** cypa <cypa!~cypam]_@5.79.187.109> has quit IRC (Ping timeout: 240 seconds)
[09:37:50] *** ypankov <ypankov!~ypankov@178.34.26.10> has joined #illumos
[09:46:17] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:2080:c4e4:938:31b9:5c4c:5941> has joined #illumos
[09:46:24] *** nbhauke <nbhauke!~hauke@55d475e1.access.ecotel.net> has joined #illumos
[10:00:39] *** nbhauke <nbhauke!~hauke@55d475e1.access.ecotel.net> has quit IRC (Quit: nbhauke)
[10:15:09] *** cypa <cypa!~cypam]_@5.79.187.109> has joined #illumos
[10:22:47] *** Teknix <Teknix!~pds@69.41.134.110> has quit IRC (Ping timeout: 260 seconds)
[10:23:39] *** Teknix <Teknix!~pds@69.41.134.110> has joined #illumos
[10:34:14] *** Teknix <Teknix!~pds@69.41.134.110> has quit IRC (Ping timeout: 272 seconds)
[10:34:51] *** Teknix <Teknix!~pds@172.58.47.49> has joined #illumos
[10:39:10] *** Teknix <Teknix!~pds@172.58.47.49> has quit IRC (Ping timeout: 246 seconds)
[10:41:56] *** cypa <cypa!~cypam]_@5.79.187.109> has quit IRC (Remote host closed the connection)
[10:47:09] *** Teknix <Teknix!~pds@69.41.134.110> has joined #illumos
[10:54:46] *** wacki <wacki!~wacki@i577A54CB.versanet.de> has joined #illumos
[11:31:43] *** trisk <trisk!~trisk@2601:196:4700:3a:8:20ff:fe3e:890e> has quit IRC (Remote host closed the connection)
[11:51:10] *** tomww <tomww!~tom@unaffiliated/tomww> has quit IRC (Quit: leaving)
[11:51:44] *** tomww <tomww!~tom@unaffiliated/tomww> has joined #illumos
[11:59:59] <richlowe> so now all I need is approximately a million years to unlearn typing 'nawk' rather than 'awk'
[12:12:42] <Woodstock> richlowe: do you know how hard it would be to make our ld grok wildcards in mapfiles?
[12:13:14] *** cypa <cypa!~cypam]_@5.79.187.109> has joined #illumos
[12:36:16] <igork> tsoome: https://paste.dilos.org/?36ad284fcba8cb4d#8XvrAA5pynfj+fslW/f+QHx6LCsa7UWIBmq6nEvdhuU=
[12:36:25] <igork> did you see something like on sparc?
[12:36:42] <igork> debug build
[12:38:21] <igork> i couldn't found where zvp defined
[12:39:36] <tsoome> http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/common/vm/seg_kmem.h#73
[12:40:54] <igork> thanks
[12:41:01] <igork> probably missmerge
[12:41:05] <tsoome> seems so
[13:02:55] <richlowe> Woodstock: I don't.
[13:05:42] <richlowe> LeftWing: I feel like banner stuff deserves a heads-up
[13:08:14] *** nbhauke <nbhauke!~hauke@55d4f653.access.ecotel.net> has joined #illumos
[13:08:17] *** xtuh_ <xtuh_!~xtuh@88.198.230.182> has joined #illumos
[13:09:26] <Woodstock> tsoome: fyi, there's a rather serious buffer overrun in your iwn change that trashes the stack so hard the system can't even take a core dump anymore.
[13:09:59] <xtuh_> Hi, i'm new to illumos, and looking for live distro that can be used for move one of to partitions to the end of the disk. Any suggestions? or maybe there is a list of the live distros?
[13:10:54] <tsoome> Woodstock huh, so the buffer size calculation must be busted.
[13:11:23] <xtuh_> also how the drive sdb will called in illumos?
[13:11:25] <tsoome> will check it later:D
[13:12:01] *** nbhauke <nbhauke!~hauke@55d4f653.access.ecotel.net> has quit IRC (Client Quit)
[13:14:19] <wacki> xtuh_: I think the only "live distro" we have is OpenIndiana.
[13:14:52] <igork> xtuh_: it depend what you are interested in - live can be prepared with unix + boot_archive
[13:15:23] <Woodstock> doesn't OI even ship (g)parted? that should work to move partitions
[13:16:01] <Woodstock> the disks and their names can be listed with diskinfo, but iirc parted uses its own naming
[13:16:22] <xtuh_> gperted is closing while it scans disks. its VM on the esxi, may it because of no vm-tools installed?
[13:16:46] <xtuh_> how to start parted /dev/sdb on illumos?
[13:17:08] <igork> is it recognise SOLARIS2 partition?
[13:17:27] <igork> with slieses
[13:17:37] <igork> slice
[13:18:21] <Woodstock> xtuh_: it may just be /dev/rdsk/cXt1d0p0, with X being any number, but then maybe parted really wants to use the block devices in /dev/dsk
[13:18:40] <Woodstock> xtuh_: run diskinfo first to see what you got
[13:18:52] <xtuh_> thanks
[13:19:51] <wacki> ytuh_: I don't understand your use of partitions if you have an EXSi host and want to run an illumos distro as a guest.
[13:21:51] <xtuh_> i already have a vm, that was setted up by another person, it has /dev/sdb2 that needed to be moved to the end of the drive(then i can grow sdb1),
[13:40:22] *** nbhauke <nbhauke!~hauke@55d428ab.access.ecotel.net> has joined #illumos
[13:41:46] <xtuh_> parted says it can not detect filesystem :(
[13:42:03] <liv3010m> andyf: because of the freeze code for r151034, the bge driver doesn't have the latest commits from Illumos gate (mostly on Apr 24th), even if there is a size difference between this one and the one on bloody-20200413.iso, right? so the driver you gave me for upcoming r34 is no supporting my adapter and that's most likely the reason the OS hanged at boot. If this is correct, then I have no means of testing the vnic problem over
[13:42:03] <liv3010m> "single" datalinks
[13:43:32] *** nbhauke <nbhauke!~hauke@55d428ab.access.ecotel.net> has quit IRC (Client Quit)
[13:47:11] *** wacki <wacki!~wacki@i577A54CB.versanet.de> has quit IRC (Read error: Connection reset by peer)
[13:50:57] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has quit IRC (Ping timeout: 265 seconds)
[13:52:22] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has joined #illumos
[13:52:24] *** kerberizer <kerberizer!~luchesar@wikipedia/Iliev> has quit IRC (Ping timeout: 265 seconds)
[13:54:44] *** kerberizer <kerberizer!~luchesar@wikipedia/Iliev> has joined #illumos
[13:58:35] <xtuh_> how to identify and move p9 to the end of disk? https://clip2net.com/s/47kecsX https://clip2net.com/s/47keg2w
[14:00:02] *** cypa <cypa!~cypam]_@5.79.187.109> has quit IRC (Remote host closed the connection)
[14:10:01] <xtuh_> ok nevermin
[14:14:00] *** hadfl <hadfl!~hadfl@omniosce.ee.ethz.ch> has quit IRC (Quit: nn o/)
[14:14:00] *** andyf <andyf!~andyf@omniosce.ee.ethz.ch> has quit IRC (Quit: nn)
[14:14:22] *** arnoldoree <arnoldoree!~arnoldore@2001:d08:2080:c4e4:938:31b9:5c4c:5941> has quit IRC (Ping timeout: 260 seconds)
[14:14:25] *** andyf <andyf!~andyf@omniosce.ee.ethz.ch> has joined #illumos
[14:14:50] *** arnoldoree <arnoldoree!~arnoldore@113.210.105.90> has joined #illumos
[14:14:51] *** nbhauke <nbhauke!~hauke@55d41425.access.ecotel.net> has joined #illumos
[14:14:55] *** hadfl <hadfl!~hadfl@omniosce.ee.ethz.ch> has joined #illumos
[14:20:19] *** nbhauke <nbhauke!~hauke@55d41425.access.ecotel.net> has quit IRC (Ping timeout: 260 seconds)
[14:46:28] <tsoome> Woodstock found the issue:) stupid me...
[14:51:53] <tsoome> Woodstock iwn_mem_read_region_4(sc, sc->errptr, buf, sizeof (buf)); needs to be iwn_mem_read_region_4(sc, sc->errptr, buf, ARRAY_SIZE(buf));
[14:57:05] <igork> tsoome:
[14:57:05] <igork> ../../sun4/cpumem-diagnosis/cmd_memerr.c:634: error: large integer implicitly truncated to
[14:57:05] <igork> unsigned type [-Woverflow]
[14:57:18] <igork> did you see it on sparc with gcc44 build?
[14:58:39] <igork> cpuid = ULONG_MAX;
[14:59:12] <igork> with 64bit build it failed in this case
[15:14:54] <tsoome_> not really, no. it is trunkation for sure. but those modueles are built 32-bit only by default.
[15:20:20] <igork> they are 32bit builds on illumos - on dilos we have 64bits
[15:31:35] *** v_a_b <v_a_b!~volker@p57A2764F.dip0.t-ipconnect.de> has left #illumos
[15:31:41] *** v_a_b <v_a_b!~volker@p57A2764F.dip0.t-ipconnect.de> has joined #illumos
[15:46:08] *** yomisei <yomisei!~void@ip4d16be28.dynamic.kabel-deutschland.de> has quit IRC (Quit: How to defend yourself against an attacker armed with a mathematician)
[15:46:13] *** yomi <yomi!~void@ip4d16be28.dynamic.kabel-deutschland.de> has joined #illumos
[15:46:37] *** yomi is now known as Guest17450
[16:50:56] *** phyre <phyre!~phyre___@78.30.23.107> has joined #illumos
[16:52:42] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[16:53:01] <KungFuJe1us> tsoome_, igork ok, I'm back
[16:53:21] <KungFuJe1us> so I do still have solaris 11.3 on here, should I create the partition table from format on there?
[16:58:29] *** igitoor <igitoor!igitur@unaffiliated/contempt> has quit IRC (Ping timeout: 246 seconds)
[17:00:45] <neirac> toastersonerson how do I pull an image from a docker registry using podadm ?
[17:01:56] <tsoome> KungFuJe1us, yes, thats good idea
[17:05:20] <KungFuJe1us> tsoome: Inquiry with page 86h failed
[17:06:02] <KungFuJe1us> oh wait, wrong command
[17:07:47] <KungFuJe1us> ok, partitioned
[17:07:56] <KungFuJe1us> ok, what's next :)
[17:08:37] <KungFuJe1us> also, is it possible to get the VTOC version of format on x86? For future reference, anyway?
[17:08:49] <KungFuJe1us> seems like a chicken and egg problem if not
[17:09:36] <KungFuJe1us> Or is there something specific about the layout format that needs to be extremely endianness aware?
[17:09:49] <tsoome> well, you can create VOC in text file (prtvtoc for instance), and write it down with fmthard
[17:10:28] <KungFuJe1us> ah this I've done when setting up mirrors, but usually just to clone the layout from one disk to the other
[17:12:17] <KungFuJe1us> ok, now what exactly do I do with this iso? Something with ".files"? Followed by an installboot?
[17:12:21] <tsoome> but yes, the data in label is endian specific.
[17:13:39] *** igitoor <igitoor!igitur@2a00:d880:3:1::c1ca:a648> has joined #illumos
[17:15:44] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[17:16:30] <toastersonerson> neirac (IRC): poadm pull or podadm create or podadm run with the docker url after that
[17:16:41] <toastersonerson> podadm pull docker://library/ubuntu
[17:17:44] <richlowe> LeftWing: so my builds are failing check_rtime because they still say @(#)SunOS Internal Development
[17:21:00] <richlowe> LeftWing: so definitely a heads-up that you absolutely must use the new nightly? or if that's always true now, maybe we should stop installing one to /opt/onbld?
[17:21:27] <andyf> I think check_rtime was updated to handle that (and some exceptions created for the closed binaries)
[17:22:29] <andyf> ah no, sorry, it does explicitly look for illumos now..That's going to bite omnios too
[17:24:09] <andyf> patch set 2 just checked for $Line =~ /^@\(#\)[[:alpha:]]/
[17:25:48] *** sjorge_be <sjorge_be!~sjorge@unaffiliated/sjorge> has joined #illumos
[17:26:34] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 246 seconds)
[17:26:35] *** sjorge_be is now known as sjorge
[17:29:26] <andyf> (I say bite OmniOS, only because we were thinking of changing RELEASE_CM/DEV_CM in our Makefile.master, but we can change our check_rtime too as we go)
[17:29:55] <andyf> richlowe - I think that has been true for a while, seems like a good idea to drop it from onbld
[17:35:00] <pmooney> yeah, it absolutely griped in my build last night (using the omnios nightly(1))
[17:35:43] <neirac> toastersonerson are you able to build from master today? I just tried and fails
[17:36:54] <neirac> toastersonerson the version of podadm I had did not have the pull command so I pull latest changes now it does not build
[17:38:33] <toastersonerson> whoops my changes are not yet uploaded.
[17:38:44] <toastersonerson> will push now.
[17:38:56] <neirac> toastersonerson also the old version fails with this using create Error: open /podhost/images/library-ubuntu.json: no such file or directory
[17:39:26] <toastersonerson> both I know are fixed on my end. just need to upload it.
[17:39:34] <neirac> toastersonerson is a great the work you are doing on this !, thanks let me know when to pull
[17:40:05] <toastersonerson> neirac (IRC): also check that you are running host-init before anything else, to create the required datasets
[17:45:30] <Mokou> https://www.bilibili.com/video/BV1NT4y137Ef/ ymksmwr rsngr
[17:52:41] <Mokou> sorry, wrong channel
[17:58:42] <toastersonerson> neirac (IRC): try it now I can build it.
[17:59:14] <toastersonerson> make sure you are using task utility to build though. It' sets the environment varaibles as required.
[18:47:24] <neirac> toastetrsonerson just tried ../../../../pkg/mod/github.com/goodhosts/hostsfile at v0 dot 0.1/hosts.go:19:53: undefined: HostsFilePath
[18:47:30] <neirac> using task install
[18:48:35] <toastersonerson> ah interesting that it did not change version for me...
[18:50:40] <toastersonerson> neirac (IRC): fixed
[18:51:48] <neirac> toastersonerson thanks now it builds!
[18:55:44] <neirac> toastersonerson it says that host has already been initialized, but fails with Error: open /podhost/images/library-ubuntu.json: no such file or directory trying to pull the ubuntu image
[18:58:25] <toastersonerson> huh do you have zfs dataset podhost and podhost/images?
[19:01:06] <neirac> toastersonerson oh, no I need to manually create them then, let me try after that
[19:01:46] <toastersonerson> you best delete the podhost dataset and it's children then initialize again. the initialisation also adds a network definition file
[19:03:52] <neirac> toastersonerson ok, I'll do that
[19:14:41] <neirac> toastersonerson I erased rpool/podhost and it's children then run podadm host-init but still I have the same error Error: open /podhost/images/library-ubuntu.json: no such file or directory
[19:16:08] <neirac> toastersonerson I erased rpool/podhost/net has the default config but in the images directory is empty
[19:16:40] <toastersonerson> /usr/local/bin/imageadm pull docker://library/ubuntu
[19:16:46] <toastersonerson> this works for me
[19:20:06] <neirac> toastersonerson oh, I was trying to run before using imageadm that was the problem now works, but running it fails with Error: open /podhost/pods/3a5eb771-5c8f-492e-b83f-a8217c080854/root/etc/nsswitch.dns: no such file or directory
[19:20:44] <toastersonerson> yes it wont download or unpack linux images ATM need to make that work again.
[19:20:59] <toastersonerson> or rather it wont download the layers of those images
[19:23:09] <toastersonerson> although it should grab them...
[19:23:10] *** nbhauke <nbhauke!~hauke@55d41425.access.ecotel.net> has joined #illumos
[19:27:35] <neirac> oh, ok I could take a look at that at night, I remember we just need the digest to download all the blobs
[19:28:06] <toastersonerson> ah I figured the problem :)
[19:28:17] <toastersonerson> on linux there is no nsswitch,dns
[19:28:28] <toastersonerson> i need to write that file by hand. welp
[19:31:43] <neirac> toastersonerson oh, right, you need not to right it when the image is a linux one, I assume. The other thing how do I create a native container?
[19:33:01] <toastersonerson> you will need to use zhe zone import command to create an image than create it same as with any container. you can also upload the image to a registry and then use it on other hosts
[19:33:11] <toastersonerson> you can also run the imageadm build command.
[19:33:30] <toastersonerson> under supportfiles there is a number of .hcl image build configurations
[19:34:12] <neirac> toastersonerson that's really awesome I'll start trying now.
[19:34:50] <toastersonerson> also I see I need support for the full linux network stack... I only have network setup support for illumos
[19:57:02] *** izaki <izaki!~quassel@unaffiliated/izaki> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
[20:01:30] *** man_u <man_u!~manu@89-92-19-81.hfc.dyn.abo.bbox.fr> has joined #illumos
[20:09:39] <LeftWing> richlowe, pmooney, andyf: Argh! Sorry about that. I'll send a heads up mail in a bit. I agree we should stop shipping nightly and bldenv in the tools proto
[20:09:55] <LeftWing> Should I file that bug?
[20:22:28] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has joined #illumos
[20:43:33] *** tsoome__ <tsoome__!~tsoome@91.209.240.229> has joined #illumos
[20:43:33] *** tsoome_ <tsoome_!~tsoome@91.209.240.229> has quit IRC (Read error: Connection reset by peer)
[20:57:20] *** TheGreekOwl <TheGreekOwl!~Arkomeda@2a02:587:4874:d500:69f1:5548:4980:af2d> has joined #illumos
[21:01:56] *** KungFuJe1us <KungFuJe1us!~adam@72.49.160.91> has quit IRC (Read error: Connection reset by peer)
[21:02:36] *** KungFuJesus <KungFuJesus!~adam@72.49.160.91> has joined #illumos
[21:03:00] *** man_u <man_u!~manu@89-92-19-81.hfc.dyn.abo.bbox.fr> has quit IRC (Quit: man_u)
[21:12:42] <ypankov> LeftWing: YES, please
[21:12:53] <LeftWing> I'm half way through typing it out haha
[21:21:05] <andyf> How about going back to the [:alpha:] check in check_rtime?
[21:22:08] <andyf> Even if we drop bldenv and nightly from onbld, people tend to build on a variety of old and new platforms.
[21:23:12] <LeftWing> https://www.illumos.org/issues/12687
[21:23:29] <wilbury> oh boi! documentation already states that bldenv and nightly should be used from src working dir, not from onbld
[21:23:55] <LeftWing> wilbury: It's difficult to force people to read the documentation
[21:25:43] <LeftWing> andyf: Well I'm about to send the HEADS UP mail that points people in the right direction
[21:27:12] <wilbury> LeftWing: heh yes, true. i'll need to get to the sections where integration of new commands/kernel modules into build is described, if at all.
[21:27:43] <wilbury> (i've been able to hook up my work into build, but a little bit of theory never killed nobody)
[21:36:08] *** BOKALDO <BOKALDO!~BOKALDO@91.105.124.91> has quit IRC (Quit: Leaving)
[21:43:33] <LeftWing> https://gist.github.com/jclulow/f589b5bf101a4f204bd99571f7ce02cf -- does this make sense?
[21:54:06] *** vgusev <vgusev!~vgusev@188.187.60.230> has joined #illumos
[21:57:49] *** vgusev_ <vgusev_!~vgusev@188.187.60.230> has quit IRC (Ping timeout: 264 seconds)
[22:35:19] *** gitomat <gitomat!~nodebot@165.225.148.18> has quit IRC (Remote host closed the connection)
[22:35:30] *** gitomat <gitomat!~nodebot@165.225.148.18> has joined #illumos
[22:37:24] <wilbury> LeftWing: yes, it does. and maybe i'd just add somewhere near the end, very brief and clear message: "Simply put: Always use bldenv and nightly in-tree." :-)
[22:38:25] <neirac> toastersonerson I'm trying imageadm import -z pkgbuild-trunk ; but errors out with Error: requires at least 1 arg(s), only received 0
[22:39:10] <toastersonerson> yes you need to give it an image url as arg :)
[22:39:22] <toastersonerson> shorthand names are also allowed
[22:39:48] <toastersonerson> something like neirac/pkgbuild/trunk
[22:40:32] <toastersonerson> Warning this will generate tgz layer of your zone. and a extracted copy of that again
[22:43:13] <LeftWing> wilbury: I've updated the gist to hopefully make that more prominent in the first paragraph. Can you take another look?
[22:48:27] <neirac> toastersonerson, for example I have a zone named pkgbuild-trunk, imageadm -z pkgbuild-trunk neira/pkgbuild/trunk is the right invocation ?
[22:48:37] <toastersonerson> yes
[22:48:50] <toastersonerson> oh wait
[22:49:00] <toastersonerson> imageadm import -z pkgbuild-trunk neira/pkgbuild/trunk
[22:50:21] *** ypankov <ypankov!~ypankov@178.34.26.10> has quit IRC (Ping timeout: 265 seconds)
[22:52:33] <neirac> toastersonerson I got Error: open pkgbuild-trunk/root: no such file or directory
[22:52:59] <toastersonerson> what brand is it?
[22:53:23] <toastersonerson> also is it's zonepath mounted?
[22:54:19] *** kamil_warszawa <kamil_warszawa!~kamil@188.146.230.17.nat.umts.dynamic.t-mobile.pl> has joined #illumos
[22:54:43] <kamil_warszawa> hi, what features of linux containers/namespaces/cgroups are missing in zones?
[22:54:45] <toastersonerson> is reproducible seems a small bug
[22:56:04] <neirac> toastersonerson Is an illumos brand
[22:58:03] <wilbury> LeftWing: clear now.
[23:01:01] <LeftWing> kamil_warszawa: Are you referring to LX zones which provide an emulated Linux ABI, or are you asking more generally for analogous features in native illumos zones?
[23:01:45] <kamil_warszawa> native zones, not branded flavors
[23:02:01] <kamil_warszawa> so features
[23:02:27] <toastersonerson> neirac (IRC): I pushed a fix for that. Build and try again
[23:02:56] <neirac> toastersonerson ok trying now
[23:02:58] <kamil_warszawa> LeftWing: so which features overal, which are specific to zones, which to linux
[23:03:10] <LeftWing> That's a pretty big question
[23:03:14] <kamil_warszawa> between the illumos and linux kernel
[23:03:23] <LeftWing> Zones and cgroups both have a lot of different features
[23:03:36] <LeftWing> Is there some particular feature you're looking for?
[23:03:53] <kamil_warszawa> let's start with namespaces
[23:04:14] <LeftWing> Well a zone is not a namespace so much as it is a limit on visibility
[23:04:33] <LeftWing> For pids at least
[23:04:41] <LeftWing> For networking it's more like a network namespace
[23:05:12] <LeftWing> For file system stuff it's less like a mount namespace and more like a high strength chroot
[23:05:24] <toastersonerson> We do have a full virtual network stack though
[23:05:55] <toastersonerson> With Switches and virtual nics
[23:06:04] <kamil_warszawa> toastersonerson: please elaboreate, does it differ to linux?
[23:06:12] <kamil_warszawa> elaborate*
[23:06:31] <toastersonerson> Our networking is more akin to OpenvSwitch
[23:06:41] <toastersonerson> It is not comparable at all
[23:07:10] <kamil_warszawa> is it similar to vnet?
[23:07:13] <toastersonerson> But our networking is much more integrated and easier to use than openvswitch
[23:07:29] <kamil_warszawa> openvswitch is afaik openflow
[23:07:40] <LeftWing> In general, zones are a cohesive set of features that attempt to create the experience of a separate machine -- but sharing the kernel instead of requiring a hypervisor. There are some limits to make that safe; e.g., you cannot load a kernel module or create your own device nodes in a zone.
[23:07:40] <Smithx10> There was a RFD about some parts of namespaces https://github.com/joyent/rfd/blob/master/rfd/0055/README.md
[23:08:17] <toastersonerson> Smithx10 (IRC): Yes but that was an addition to make systemd work
[23:08:31] <toastersonerson> Not a full mount namespace
[23:08:41] <Smithx10> gotcha
[23:08:44] <LeftWing> Mount namespaces are somewhat questionable in general
[23:09:21] <toastersonerson> kamil_warszawa (IRC): I do not know vnet. and a wuick google search shows me Azure and GCP only. Do you have a link?
[23:09:29] <LeftWing> They present a lot of observability challenges; e.g., if each process can have a totally different view of the VFS, it can be hard to know what a stat() call or an open() call in a particular process will end up with
[23:09:35] <kamil_warszawa> toastersonerson: freebsd vnet
[23:11:11] <toastersonerson> kamil_warszawa (IRC): I think we have more features. But basicly yes
[23:13:37] <kamil_warszawa> what features in zones are missing if we compare to linux container apis
[23:13:42] <LeftWing> As with all these things, I think it helps to evaluate options once you have a specific workload in mind
[23:13:43] <kamil_warszawa> mount namespace is one
[23:14:33] <toastersonerson> Our emulation API is mostly missing the linux COntainer API's. As we have no concept of nested zones.
[23:15:10] <toastersonerson> And it is questionable if we want those API's useing them inside COntainers is terrible practice even on linux.
[23:15:23] <kamil_warszawa> is it somewhere hardcoded in the design of zones?
[23:15:34] <kamil_warszawa> can it be simply expanded to add nested zones?
[23:15:45] <LeftWing> It's all just software
[23:16:04] <LeftWing> If we need nested zones for something, I am sure it could be done
[23:16:45] <toastersonerson> It's more the question can we make a proper design that makes sense in a majority of cases. Nesting is also in Freebsd and usefull there
[23:16:48] <LeftWing> I don't think it would be a small body of work, though
[23:17:18] <kamil_warszawa> if we would start over with zones, what would we do differently
[23:17:25] <toastersonerson> Question of course how Ressource limitations work in that case. A flat structure makes things a lot easier and stable
[23:17:27] <kamil_warszawa> nesting is now
[23:17:41] <kamil_warszawa> wanted
[23:17:50] <LeftWing> I'm not completely sure that we would add nesting if we designed it again
[23:18:17] <LeftWing> It might be useful in some cases, but it also makes it much harder to see what's going on in the system when you consider the management tools and such
[23:18:35] <LeftWing> Today, there is a column in ps and prstat and so on that shows you which zone a process is in
[23:18:45] <toastersonerson> As for redesigning. mostly just make the tooling have a Rust API and not use XML :)
[23:18:51] <LeftWing> Without nesting, it's easy to understand that that column means
[23:19:10] <LeftWing> If you were to nest, what would you show in that column for the operator in the global zone
[23:19:30] <LeftWing> Would you show the top-level zone, or would you break it down into child zones, etc
[23:20:52] <kamil_warszawa> in ps I would show the internal zone
[23:21:05] <Smithx10> I'm not to sure I understand the use case
[23:21:08] <LeftWing> Right, but that internal zone name would be chosen by the zone manager
[23:21:10] <kamil_warszawa> with an option to show nested view, like procstat
[23:21:12] <LeftWing> Not by the global operator
[23:21:37] <LeftWing> So it could be at best questionable, if not downright misleading
[23:21:53] <LeftWing> Also do you allow nested zones to have the same name as other zones (nested or otherwise) on the system?
[23:21:57] <LeftWing> Today we do not allow duplicates
[23:21:58] <kamil_warszawa> is there need to keep name unique?
[23:22:08] <LeftWing> Yes because lots of the APIs refer to zones by name
[23:22:14] <toastersonerson> Yes
[23:22:27] <kamil_warszawa> why not refer to uuid?
[23:22:37] <toastersonerson> although you are allowed to use a UUID as a name like smartos or my container engine does
[23:22:45] <LeftWing> Right, you can choose to use UUIDs for names
[23:22:47] <LeftWing> If you want
[23:22:58] <LeftWing> But you can also choose the more manageable option of names, if that suits you
[23:23:16] <LeftWing> We've tried to create a mechanism rather than a policy there
[23:23:25] <kamil_warszawa> what other changes in design would be done if we would start over
[23:24:06] <LeftWing> I'm not sure. I'd recommend using the system for something to get some feel for its strengths and weaknesses before trying to evaluate it in the abstract.
[23:24:35] <kamil_warszawa> bug generally it feels like feature complete, am I right?
[23:24:45] <toastersonerson> yep
[23:25:20] <LeftWing> I think we've got a pretty good base, and one which has supported a variety of incremental improvements over ~15 years
[23:25:36] <toastersonerson> I am working on integrating it with the CNCF toolchain right now. but that is added on top not in the design of zones itself
[23:26:08] <LeftWing> We can certainly add new features where it makes sense, and is appropriately designed, reviewed, tested, etc. If you use the software and find something missing, we can help you work on it.
[23:26:10] <kamil_warszawa> do they integrate well with OCI / Docker?
[23:26:12] <kamil_warszawa> I think so
[23:26:20] <toastersonerson> It is OCI :)
[23:26:34] <toastersonerson> and Yes you can already push/pull and run docker images :)
[23:26:39] <toastersonerson> native ones at least
[23:26:46] <toastersonerson> working on making the lx brand work
[23:26:52] <kamil_warszawa> native ones are enough
[23:27:06] <LeftWing> Docker's use of Linux containers ends up looking a lot like zones, in the set of features they choose to expose and the layout of their container
[23:27:08] <toastersonerson> https://git.wegmueller.it/opencloud/opencloud
[23:27:08] <Smithx10> toastersonerson: do you have this written up anywhere?
[23:27:18] <toastersonerson> only in the readme thus far
[23:28:18] <LeftWing> Some other container stuff in, I suppose the Kubernetes ecosystem, make different choices about overlaps in different namespaces, which doesn't fit in as directly with the zones model
[23:28:24] <kamil_warszawa> does layering work with zones?
[23:28:33] <kamil_warszawa> like union fs mount
[23:28:45] <toastersonerson> yes that is implemented in ZFS with snapshots
[23:28:52] <toastersonerson> same as the docker ZFS driver
[23:28:57] <kamil_warszawa> overlay mount
[23:29:01] <toastersonerson> you do not need union fs
[23:29:09] <toastersonerson> its optional on linux
[23:29:43] <sjorge> Anyone using LDAP... also using the native smb/server ?
[23:30:00] <sjorge> Anyway to tell it to pull the passwords from ldap too? it doesn't seem to do so
[23:31:28] <LeftWing> We don't have a native union file system -- they present some complex semantic challenges
[23:32:05] <LeftWing> We do have something like bind mounts though, which we call "lofs"
[23:32:30] <LeftWing> As toastersonerson notes, we can get a lot of what you might get with overlay mounts with ZFS snapshots and clones instead
[23:32:48] <Smithx10> kamil_warszawa: what problem are you trying to solve ? What is your use case for the sysem?
[23:33:09] <toastersonerson> Actually overlay fs was a workaround of docker for linux distros which do not have btrfs or ZFS it is the preffered methof on linux
[23:33:31] <toastersonerson> ZFS or btrfs that is
[23:34:24] <LeftWing> Yeah, ZFS (or whatever snapshotting system) avoids the semantic complexity of the unionfs approach
[23:34:26] <kamil_warszawa> Smithx10: I'm trying to research this field
[23:34:52] <LeftWing> Research for a specific use case or project, or is this like... academic research?
[23:35:25] <kamil_warszawa> for now we can call it selfeducational one; I'm evaluating if I would start over in a different kernel what would I do differently if I would pick zones as base
[23:35:43] <kamil_warszawa> and how much time I would need to spend on it
[23:35:46] <Smithx10> You are building temple_os?
[23:35:48] <kamil_warszawa> 1 year?
[23:35:53] <kamil_warszawa> no
[23:35:56] <Smithx10> hahha that was a joke :P
[23:36:40] <wilbury> kamil_warszawa: tired from netbsd?
[23:36:51] <kamil_warszawa> no tired!
[23:37:11] <kamil_warszawa> but tired from chroot
[23:37:16] <LeftWing> I think if you want to seriously evaluate this stuff you'll want to go and install it and make it do something
[23:37:18] <wilbury> ah, yes.
[23:37:42] <LeftWing> And have a look in the source. There are some good design comments lying around. There is also at least one paper on Zones, I believe.
[23:38:04] <kamil_warszawa> I already inspected some of the source code
[23:38:17] <kamil_warszawa> but paper on zones hmm
[23:38:23] <kamil_warszawa> it would be interesting
[23:38:43] <toastersonerson> it's based on the FreeBSD Jail paper and extends it from there
[23:38:48] <igork> zones design are fine with parallel access to the same space
[23:38:59] <toastersonerson> Some real engineering work from Sun Microsystems back in the day
[23:39:07] <kamil_warszawa> toastersonerson: link? or name of it?
[23:39:25] <toastersonerson> I do not have it sorry.
[23:39:39] <igork> toastersonerson: zones are more flexible
[23:39:40] <kamil_warszawa> https://www.usenix.org/legacy/event/lisa04/tech/full_papers/price/price.pdf ?
[23:40:08] <wilbury> also PSARC might be helpful
[23:40:15] <wilbury> (to see original intentions)
[23:44:05] <kamil_warszawa> wilbury: playstation archive?
[23:45:08] <kamil_warszawa> toastersonerson: OCI has a command to kill() a container
[23:45:20] <kamil_warszawa> does it work like sending any signal to all processes within a container?
[23:45:42] <kamil_warszawa> and how is it implemented with zones
[23:46:02] <toastersonerson> we have halt which cleans it with kills it from kernel. No idea what they do on linux
[23:46:13] <toastersonerson> I work with the standards not the implementation
[23:46:23] <toastersonerson> the standards have no kill
[23:47:12] <kamil_warszawa> toastersonerson: https://github.com/opencontainers/runtime-spec/blob/e09c7c488e298f2147734e657eba72fc9fa10ca9/runtime.md#kill
[23:47:19] <kamil_warszawa> kill <container-id> <signal>
[23:47:45] <wilbury> kamil_warszawa: https://illumos.org/opensolaris/ARChive/PSARC/
[23:48:11] <toastersonerson> ah. I have mapped that to zoneadm halt
[23:48:40] <kamil_warszawa> toastersonerson: ah, are you the person who added oci support?
[23:48:50] <toastersonerson> yes
[23:48:57] <toastersonerson> oh no
[23:49:04] <toastersonerson> not in oci that was oracle
[23:49:16] <kamil_warszawa> runc
[23:49:22] <neirac> toastersonerson ok now it worked
[23:49:32] <toastersonerson> runc has support for zones?
[23:50:29] <kamil_warszawa> toastersonerson: https://github.com/guillermomolina/solaris-oci
[23:50:40] <toastersonerson> I made my own engine which implements the specs
[23:50:48] <kamil_warszawa> toastersonerson: link?
[23:50:55] <toastersonerson> https://git.wegmueller.it/opencloud/opencloud
[23:52:37] <kamil_warszawa> is it functional now?
[23:53:01] <toastersonerson> yes
[23:53:08] <toastersonerson> not finished though
[23:53:16] <toastersonerson> need to fix some bugs :)
[23:54:59] <neirac> toastersonerson I tried to create a container using hte image
[23:55:03] <neirac> but failed https://pastebin.com/hGGJf6YB
[23:55:09] <kamil_warszawa> toastersonerson: I will /query you
[23:56:04] <toastersonerson> neirac (IRC): what does ls /podhost/images/ say?
[23:56:23] <toastersonerson> is the json file there?
[23:57:42] <neirac> toastersonerson my fault, the url provided was wrong now it's created, is really fast to provision
[23:58:30] <toastersonerson> Yeah zfs clone :)
[23:58:42] <neirac> toastersonerson oh no, trying to run the image failed
[23:59:14] <toastersonerson> neirac (IRC): what was wrong with the url?
[23:59:25] <toastersonerson> ah yes I see
[23:59:28] <toastersonerson> no C :)
top

   May 2, 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 | >