Switch to DuckDuckGo Search
   November 18, 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 | >


NOTICE: This channel is no longer actively logged.

Toggle Join/Part | bottom
[00:17:45] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has quit IRC (Quit: Leaving)
[00:21:54] *** andy_js <andy_js!~andy@94.11.180.149> has quit IRC (Quit: andy_js)
[00:40:36] *** tomww <tomww!~tom@unaffiliated/tomww> has joined #illumos
[00:54:13] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-ucuqvppoefjbqqnf> has quit IRC (Ping timeout: 265 seconds)
[00:54:54] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-rzvztrwoyjjvgkld> has joined #illumos
[01:13:13] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[01:15:54] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[01:30:07] <Reinhilde> if someone wants to implement an RBAC system that would be theoretically compatible with illumos' into another operating system, and doesn't want to get tarpitted by the CDDL, is it mandatory that they throw away their copy of the gate, undergo deprogramming rituals to forget everything they read, and replicate the behavior solely from manpages?
[01:46:42] <arekinath> if you really need a reliable answer to that, ask a lawyer who specialises in software copyright about what the standards are for when something is or isn't a derived work
[01:46:52] <arekinath> but it'll probably be a very involved answer
[01:47:31] <arekinath> the internet is likely to give you conflicting answers
[01:48:52] <Reinhilde> i don't need a reliable answer arekinath
[01:49:23] <Reinhilde> I just need a 95th percentile "I don't think anyone would sue over this"
[01:49:24] <LeftWing> Then the answer is "Chesterfield Sofa"
[01:49:33] <arekinath> then go ask on twitter and you'll get all the armchair lawyers you need
[01:50:29] <Reinhilde> so I should implement my own incompatible RBAC system
[01:50:44] <arekinath> I'm not sure what you would hope to gain by trying to be compatible?
[01:51:18] <Reinhilde> familiarity of the APIs if someone wanted to use it in their own program
[01:51:18] <arekinath> not an awful lot of code outside of the gate relying on the solaris/illumos privs and rbac infra
[01:51:35] <arekinath> and not that many people with high familiarity with it tbh
[01:51:36] <arekinath> haha
[01:51:46] <Reinhilde> which is like, I really don't mind anything to make developers' lives easier
[01:52:42] <arekinath> I suspect you would probably do better to take inspiration from it, and also some inspiration from other things like openbsd's pledge()/unveil() etc and think up an API that fits with whatever OS it is you're planning to enhance
[01:53:30] <Reinhilde> degrade, not enhance, arekinath ;-)
[01:54:00] <Reinhilde> I'm also planning to write the second stage init in Tcl and have the directory structure more reminiscent of Windows than of any UNIXoid programme
[01:54:27] <arekinath> ok that's nice. I feel like that doesn't have much to do with #illumos
[01:54:47] <Reinhilde> it does not.
[01:54:55] <Reinhilde> however, the matter of RBAC does
[01:55:20] <Reinhilde> what pages from sections of the illumos manual book do I need to send to my typesetter
[01:55:38] <Reinhilde> (to be legit, I'm trolling and don't expect an answer, because I could probably get one from apropos(1))
[01:57:35] <LeftWing> Had any stacks correlate with a SIGBUS yet?
[01:58:29] <Reinhilde> LeftWing: No.
[01:59:41] <LeftWing> Is that because you also haven't had any further SIGBUS issues?
[02:00:02] <Reinhilde> LeftWing: it's because I haven't put enough weight on the machine to get any further Bus Error issues.
[02:00:06] <LeftWing> Ok
[02:07:26] <Cthulhux> "tarpitted by the CDDL" lol
[02:07:31] <Cthulhux> one day
[02:07:44] <Cthulhux> i'll write a book about "shit non-CDDLers say"
[02:08:36] <Reinhilde> Cthulhux: lol
[02:08:45] <Reinhilde> i guess it's not as thick of a tarpit as the GPL.
[03:23:02] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[03:23:37] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[03:37:27] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[04:17:52] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[04:46:42] *** AllanJude <AllanJude!~allan@freebsd/developer/AllanJude> has quit IRC (Ping timeout: 265 seconds)
[04:49:17] *** AllanJude <AllanJude!~allan@freebsd/developer/AllanJude> has joined #illumos
[05:22:13] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has quit IRC (Remote host closed the connection)
[05:31:06] *** wl_ <wl_!~wl_@2605:6000:1b0c:5a5f::904> has joined #illumos
[05:39:24] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has joined #illumos
[05:41:53] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[06:37:03] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has quit IRC (Remote host closed the connection)
[07:16:26] *** Diagon <Diagon!DiagonalAr@gateway/vpn/protonvpn/diagonalarg> has joined #illumos
[07:29:51] *** Diagon <Diagon!DiagonalAr@gateway/vpn/protonvpn/diagonalarg> has quit IRC (Quit: Leaving)
[07:34:14] *** Diagon <Diagon!DiagonalAr@gateway/vpn/protonvpn/diagonalarg> has joined #illumos
[07:46:19] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Quit: amrfrsh)
[07:57:40] *** snuff-work <snuff-work!~snuff-wor@202-161-112-134.tpgi.com.au> has quit IRC (Quit: Leaving.)
[08:12:32] *** amrfrsh <amrfrsh!~Thunderbi@131.234.44.76> has joined #illumos
[08:14:16] *** Reinhilde is now known as AmyMalik
[08:19:06] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has joined #illumos
[08:29:58] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: tsoome)
[08:36:48] *** joltman_afk <joltman_afk!znc@gateway/vpn/privateinternetaccess/joltman> has joined #illumos
[08:37:02] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[08:38:44] *** joltman <joltman!~znc@unaffiliated/joltman> has quit IRC (Ping timeout: 276 seconds)
[08:38:44] *** joltman_afk is now known as joltman
[08:43:42] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has quit IRC (Remote host closed the connection)
[08:44:09] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has joined #illumos
[09:16:33] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[09:39:52] *** az0r <az0r!~sh42@fsf/member/zeroSignal> has quit IRC (Quit: leaving)
[09:40:18] *** BH23 <BH23!~BH23@193.117.206.132> has joined #illumos
[09:48:56] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[09:56:29] *** andy_js <andy_js!~andy@94.11.180.149> has joined #illumos
[10:10:32] *** snuff-work <snuff-work!~snuff-wor@202-161-112-134.tpgi.com.au> has joined #illumos
[10:33:56] *** BH23 <BH23!~BH23@193.117.206.132> has quit IRC (Quit: Leaving)
[10:34:22] *** BH23 <BH23!~BH23@193.117.206.132> has joined #illumos
[10:35:42] *** _Joes_ <_Joes_!~BH23@193.117.206.132> has joined #illumos
[10:58:24] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[11:08:24] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has joined #illumos
[11:10:03] *** amrfrsh <amrfrsh!~Thunderbi@131.234.44.76> has quit IRC (Quit: amrfrsh)
[11:21:55] *** ngchk1 <ngchk1!~ngchk1@b2b-92-50-91-166.unitymedia.biz> has joined #illumos
[11:37:30] *** andy_js_ <andy_js_!~andy@94.11.180.149> has joined #illumos
[11:40:52] *** andy_js_ <andy_js_!~andy@94.11.180.149> has quit IRC (Client Quit)
[11:42:02] *** andy_js_ <andy_js_!~andy@94.11.180.149> has joined #illumos
[11:42:43] *** andy_js_ <andy_js_!~andy@94.11.180.149> has quit IRC (Client Quit)
[11:54:17] <clapont> hi, everyone!
[11:55:42] *** andy_js_ <andy_js_!~andy@94.11.180.149> has joined #illumos
[11:57:38] *** andy_js <andy_js!~andy@94.11.180.149> has quit IRC (Ping timeout: 276 seconds)
[11:57:38] *** andy_js_ is now known as andy_js
[11:59:50] <tsoome> hi
[12:01:00] *** andy_js_ <andy_js_!~andy@94.11.180.149> has joined #illumos
[12:04:14] *** pwinder <pwinder!~pwinder@86.2.210.254> has joined #illumos
[12:19:43] *** andy_js_ <andy_js_!~andy@94.11.180.149> has quit IRC (Quit: andy_js_)
[12:39:18] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[13:02:57] *** zsj <zsj!~zsj@BC0607BE.catv.pool.telekom.hu> has quit IRC (Remote host closed the connection)
[13:06:22] *** zsj <zsj!~zsj@BC0607BE.catv.pool.telekom.hu> has joined #illumos
[13:11:05] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has quit IRC (Quit: Leaving)
[13:13:40] *** zsj <zsj!~zsj@BC0607BE.catv.pool.telekom.hu> has quit IRC (Quit: leaving)
[13:29:55] *** bacterio <bacterio!bacterio@fsf/member/bacterio> has quit IRC (Read error: Connection reset by peer)
[13:39:03] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[13:39:39] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[14:20:26] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has joined #illumos
[14:28:10] *** baojg_ <baojg_!~baojg@2604:a880:0:1010::16:1> has joined #illumos
[14:29:44] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has quit IRC (Ping timeout: 276 seconds)
[14:35:28] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has joined #illumos
[14:36:38] *** baojg_ <baojg_!~baojg@2604:a880:0:1010::16:1> has quit IRC (Ping timeout: 252 seconds)
[14:51:22] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:53:47] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has quit IRC (Ping timeout: 250 seconds)
[15:33:08] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[15:33:12] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[15:52:53] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Ping timeout: 252 seconds)
[15:54:41] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 246 seconds)
[16:00:02] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has quit IRC (Ping timeout: 240 seconds)
[16:18:56] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[16:19:37] *** Diagon <Diagon!DiagonalAr@gateway/vpn/protonvpn/diagonalarg> has quit IRC (Ping timeout: 240 seconds)
[16:24:47] *** tomww <tomww!~tom@unaffiliated/tomww> has quit IRC (Ping timeout: 246 seconds)
[16:30:02] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 240 seconds)
[16:32:29] *** psarria <psarria!~psarria@26.red-79-146-96.dynamicip.rima-tde.net> has quit IRC (Read error: Connection reset by peer)
[16:33:10] *** Diagon <Diagon!DiagonalAr@gateway/vpn/protonvpn/diagonalarg> has joined #illumos
[16:34:27] *** tomww <tomww!~tom@unaffiliated/tomww> has joined #illumos
[16:34:45] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[16:38:46] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[17:06:31] *** pwinder <pwinder!~pwinder@86.2.210.254> has quit IRC (Quit: This computer has gone to sleep)
[17:16:38] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[17:23:21] <gitomat> [illumos-gate] 11855 Update SMB client to 3.02 -- Gordon Ross <gwr at nexenta dot com>
[17:26:32] <KungFuJesus> tsoome: sorry I dropped off the face of the earth after Saturday, were you able to make any progress toward that bug?
[17:27:45] <KungFuJesus> I was thinking a good debug measure might be to shim in loop detection when traversing that vdev linked list to find exactly which nodes might be causing the stall
[17:28:33] <KungFuJesus> I mean, it sure seems like it's an infinite loop, but given that stage of the boot process there's very little protecting memory, so it could be a more insideous memory corruption bug
[17:29:44] <tsoome> I have this nasty feeling that we get some sort of side effect there - but we will need to print out more data, just by reading the code I was not able to see anything unusual. which means we must have some bad value
[17:30:25] <KungFuJesus> at least one other person in this channel had that problem, I don't recall the user name but my solution to get booting worked for them
[17:30:37] <KungFuJesus> (disabling the SAS controller from presenting disks to the BIOS)
[17:31:52] <KungFuJesus> I guess the bright side to this is the boot loader is a short lived process, so once it successfully inits the kernel any bug that corrupts memory won't have other adverse effects. It's just getting past the bootloader that becomes the problem :)
[17:31:55] <tsoome> I know about 2 more reports. anyhow, I’ll finish with other thing first, then we can start to iterate..
[17:43:57] *** pwinder <pwinder!~pwinder@86.2.210.254> has joined #illumos
[17:45:48] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has joined #illumos
[17:50:18] *** zsj <zsj!~zsj@BC0607BE.catv.pool.telekom.hu> has joined #illumos
[18:06:53] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has joined #illumos
[18:17:52] <v_a_b> tsoome I am having trouble booting OOCE 032 -- over in #omnios, andyf told me to bother you :-)
[18:18:09] <tsoome> hm?
[18:18:26] <tsoome> troube in which way?
[18:18:38] <v_a_b> I can boot OOCE 028 and 030 off a USB key just fine, but the loader doesn't even start for 032.
[18:18:51] <v_a_b> Being desperate I dd'ed the iso image to the USB key. Same result.
[18:19:00] <v_a_b> 030 ISO boots fine, 032 hangs.
[18:19:04] <tsoome> any messages?
[18:19:09] <v_a_b> The loader doesn't even start. No messages.
[18:19:22] <v_a_b> I am still in BIOS so ctrl-alt-del works and causes a reboot.
[18:19:26] <tsoome> nothing at all? just dark screen?
[18:19:45] <v_a_b> The usual "empty" BIOS screen with the cursor in the upper left corner.
[18:20:09] <v_a_b> It's actually the same HP Microserver G7 that we did the PXE tests on many months ago. :-)
[18:21:23] <v_a_b> It's completely reproducible: dd 030, boot, works, dd 032, boot, hangs
[18:21:53] <v_a_b> Don't know how to debug this. :-(
[18:22:46] <tsoome> bios or uefi boot?
[18:22:59] <v_a_b> The G7 is old, BIOS only.
[18:23:47] <v_a_b> I played around with the "USB device emulation", changing it from "auto" to "harddisk" and "CD-ROM", no difference.
[18:24:04] <v_a_b> Next test will be using OOCE bloody.
[18:25:21] <tsoome> ok, no output at all is rather weird, it means the system failed probing even first disk
[18:26:14] <v_a_b> Or, maybe it thinks it needs to talk to ttya. Unfortunately, there is no COM interface.
[18:27:49] <v_a_b> It must be some change from 030 to 032... it seems there were lots of loader changes.
[18:28:21] <tsoome> you can use installboot -i to check the version of gptzfsboot in that build, so we would know about the state, but I’d start with checking the latest version.
[18:29:53] <v_a_b> I can boot 028 from the SSD currently in the machine. Will the installboot binary in that release show me the version of gptzfsboot on a 032 USB device?
[18:31:01] <tsoome> hm. the serial may be good lead, but we do default on video … unless something else is done in omnios.
[18:31:36] <tsoome> it may. check istallboot -i ..
[18:32:45] <v_a_b> # installboot -Vni /dev/rdsk/c2t0d0p0
[18:32:45] <v_a_b> Dry run requested. Nothing will be written to disk.
[18:32:45] <v_a_b> installing loader to whole disk device is not supported
[18:32:51] <v_a_b> open: No such file or directory
[18:32:51] <v_a_b> Failed to detect file system type
[18:32:51] <v_a_b> Unable to gather device information from /dev/rdsk/c2t0d0p0
[18:34:50] <v_a_b> I tried to be smart and specified the image file itself:
[18:34:54] <v_a_b> Unable to find multiboot header
[18:34:54] <v_a_b> Error reading bootblock from omniosce-bloody-20191018.usb-dd
[18:37:22] <tsoome> you need to use either file itself or slice with root file system
[18:37:55] <v_a_b> Just booted OOCE 133 aka bloody, same result: hangs.
[18:38:18] <tsoome> without any messages?
[18:38:25] <v_a_b> yes.
[18:38:44] <v_a_b> Summary: 028 + 030 work, 032 + 033 hang.
[18:38:53] <v_a_b> On the same physical USB key.
[18:39:09] <tsoome> in that case, just grab the source and start inserting printfs to see where..
[18:40:22] <v_a_b> OK, I'll talk to andyf when he's back, maybe he can help me with that.
[18:40:49] <v_a_b> For now, thanks!
[18:41:21] <tsoome> you need usr/src/boot/sys/boot/i386/gptzfsboot/zfsboot.c
[18:41:33] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 245 seconds)
[18:41:42] <v_a_b> OK, noted.
[18:42:16] <tsoome> the main is there and first thing the main() does is calling i386/libi386/biosdisk.c bd_init() to probe the disks
[18:46:47] <tsoome> v_a_b: can you test oi or freebsd current?
[18:47:35] <v_a_b> Good idea. I'll have a look at the latest OI USB media.
[18:48:14] <v_a_b> I have OI-hipster-minimal-20191106.usb -- going to try that now.
[18:49:04] *** jellydonut <jellydonut!~quassel@s91904424.blix.com> has quit IRC (Quit: jellydonut)
[18:52:54] <v_a_b> ... boots fine! So the problem is specific to OOCE.
[18:53:57] *** jellydonut <jellydonut!~quassel@s91904425.blix.com> has joined #illumos
[18:54:10] <tsoome> maybe the build?
[18:55:31] <tsoome> altho it should get linked with i386/boot.ldscript
[18:55:52] <v_a_b> Maybe. I'll see what andyf says. I have used the stock images from the OOCE d/l site for both 032 and bloody.
[18:56:51] <tsoome> we can check the disasm :D
[18:56:55] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[18:58:22] <v_a_b> I'm here and available; andyf will be back later. If you have a cmd for me to run on the USB image I'll be glad to do so.
[18:58:57] <tsoome> well, you can pick loader binaris from oi /boot :)
[18:59:05] <tsoome> binaries*
[18:59:51] <v_a_b> Yes, I could. Maybe the OOCE 030 loader could also load the 032 kernel. The goal is to set up a "clean" QA machine, though.
[19:00:33] <tsoome> it most likely does - the kernel startup is not really changing much
[19:00:40] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[19:04:57] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Ping timeout: 240 seconds)
[19:04:58] <tsoome> if I have chain of patches, one depending on previous, can I set up the followup(s) on gerrit?
[19:09:26] <jlevon> if you push the chain I think gerrit opens one CR for each commit. but never really tried that myself
[19:10:20] <jbk> the older versions did
[19:10:26] <v_a_b> tsoome A strings on the OOCE USB image shows 1.1-2019.10.03.1 as the loader version in 032. In bloody, it is 1.1-2019.10.09.1. These don't work. The hipster image has 1.1-2019.10.29.1 which works. So it really must be the OOCE build.
[19:10:34] <jbk> though with the older versions, you had to be careful how you updated
[19:10:57] <jbk> but I wonder if you set unique change-ids to each, that might avoid that
[19:11:06] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[19:12:58] <tsoome> ou, it actually was working
[19:28:45] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has joined #illumos
[19:34:28] *** BH23 <BH23!~BH23@193.117.206.132> has quit IRC (Ping timeout: 245 seconds)
[19:34:29] *** _Joes_ <_Joes_!~BH23@193.117.206.132> has quit IRC (Ping timeout: 246 seconds)
[19:37:48] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has quit IRC (Ping timeout: 245 seconds)
[19:42:04] *** mgerdts <mgerdts!~textual@2600:6c44:c7f:ec89:e579:67e7:835a:ada8> has quit IRC (Read error: Connection reset by peer)
[19:47:01] *** mgerdts <mgerdts!~textual@2600:6c44:c7f:ec89:d935:3dd0:f2b6:15da> has joined #illumos
[19:49:22] *** dgtlcmo <dgtlcmo!~dgtlcmo@c-24-61-203-74.hsd1.ma.comcast.net> has quit IRC (Quit: Lost terminal)
[19:55:56] <v_a_b> tsoome Same problem on a different Microserver G8, also another user reported USB boot failure on a DL380 G7 in #omnios.
[19:57:51] <tsoome> :(
[19:58:13] *** pwinder <pwinder!~pwinder@86.2.210.254> has quit IRC (Quit: This computer has gone to sleep)
[19:58:15] <v_a_b> Guess we'll have to build a loader with lots of printf()s ...
[19:58:47] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has joined #illumos
[20:04:08] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[20:05:03] <Kurlon> I don't know if it applies in this case, but I've dealt with a bunch of HP products, up to and including G6 series servers that are VERY finicky about what USB options they'll boot. I suspect it's MBR vs GPT that throws them for a loop but haven't dug too deeply.
[20:06:45] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[20:07:41] *** pwinder <pwinder!~pwinder@86.2.210.254> has joined #illumos
[20:17:37] <andyf> I’m still out but v_a_b have you tried dd-ing the iso to the usb stick? it won’t all work but it will be useful to see if it gets further.
[20:20:19] *** cpk <cpk!~chris@185.172.87.163> has joined #illumos
[20:22:13] <gitomat> [illumos-gate] 11820 upstream Nexenta iSCSI and COMSTAR fixes -- Rick McNeal <rick.mcneal at nexenta dot com>
[20:24:23] <v_a_b> andyf Yes I have, same result. 030 iso boots, 032 doesn't.
[20:26:10] <tsoome> it has to be either the source version or something about the build - binutils?
[20:26:41] *** zsj <zsj!~zsj@BC0607BE.catv.pool.telekom.hu> has quit IRC (Quit: leaving)
[20:28:01] <tsoome> but tbh, it takes a bit of checking to see what exactly. but if the error is so early we do not get single char on screen, it should not be too hard to find.
[20:28:49] *** zsj <zsj!~zsj@BC0607BE.catv.pool.telekom.hu> has joined #illumos
[20:38:07] *** pwinder <pwinder!~pwinder@86.2.210.254> has quit IRC (Quit: This computer has gone to sleep)
[20:51:27] *** bacterio <bacterio!bacterio@fsf/member/bacterio> has joined #illumos
[20:54:03] <richlowe> bisect loader, if that doesn't work build it with the old binutils?
[20:54:10] <richlowe> ... though loader changes a lot, so maybe try the old binutils/compiler first?
[20:54:38] <andyf> I'm not sure if binutils has even changed between omnios r30 and r32
[20:54:56] <andyf> I'm going to check first if I can boot it on any of my machines
[21:05:24] <tsoome> it was just an guess anyhow
[21:12:04] <andyf> I just looked for differences in the disk image layout - r32 is missing the VBR block
[21:12:05] <andyf> https://paste.ec/paste/PN5CElcp#91L-2kubGhsjDTz79g+7U3by/A3gb2NqAiHs2Lxfwp5
[21:16:57] <tsoome> *that* can explain
[21:17:19] <andyf> :)
[21:18:01] <tsoome> but it is easy check, use mdb /dev/dsk/c8t1d0p0, ::load disk_label, then ::mbr and ::gpt or ::vtoc
[21:18:57] <tsoome> just to confirm installboot
[21:19:25] <andyf> https://paste.ec/paste/wueVj5SD#2hOP3wDPj-YT67UM/gaZSvj0B425H+FSya1sT632P5P
[21:19:30] <andyf> this is one I just built on r32
[21:20:07] <andyf> so I need to check why the VBR is not being written
[21:20:24] <andyf> (cool mdb module :) )
[21:20:57] <tsoome> yes, and that does explain why there is just blank screen
[21:22:23] <tsoome> if you run installboot now, does it install vbr?
[21:23:45] <andyf> ah.. our build usb script calls installboot with s2 - if I run it with s0 then it does what I expect
[21:25:47] <tsoome> phew:)
[21:26:02] <tsoome> I was already checking if I have messed something up in installboot:D
[21:26:08] <andyf> it apparently always worked before some recent change
[21:26:16] <andyf> well, recent == last 6 months
[21:31:58] <andyf> ok, something is strange here
[21:32:10] <andyf> I have to install to s2, and then to s0 to get the right result
[21:33:36] <tsoome> you have ufs on s2?
[21:33:41] <andyf> yes
[21:34:15] <tsoome> it should install to s1
[21:34:20] <andyf> (well, that's what the script says)
[21:35:23] <tsoome> ok, gptzfsboot to s1, pmbr can go both s1 or s2, both have space for it
[21:35:29] *** bahamas10 <bahamas10!~dave@cpe-72-231-187-25.nycap.res.rr.com> has quit IRC (Ping timeout: 268 seconds)
[21:36:10] <tsoome> ok, what you get if you run installboot /dev/rdsk/c…s2 ?
[21:36:27] <tsoome> you will probabluy need -mf too
[21:36:36] <andyf> https://paste.ec/paste/CYr6kPHc#chKTcKQVJHxSUsTRTeieup7vBwtZaK0CeawATQdvmqJ
[21:36:49] <andyf> sorry about the prompt making it a bit harder to read
[21:37:21] <tsoome> ok, there was no pcfs in s0
[21:37:48] <tsoome> gptzfsboot was writter there: bootblock written for /dev/rdsk/c3t1d0s1, 273 sectors starting at 1 (abs 69889)
[21:39:10] <tsoome> and stage1 written to slice 1 sector 0 (abs 69888) - thats s1
[21:39:39] <tsoome> the installboot does always need slice with yout root file system
[21:40:37] <andyf> that's slice2 here, UFS
[21:41:10] <tsoome> yes.
[21:41:44] <tsoome> once you have pcfs on s0, the installboot does accept it (we support boot ing from pcfs for smartos)
[21:41:45] <andyf> and it formats the ESP ok
[21:42:11] <andyf> ok, but there is no VBR after installboot ... s2
[21:42:56] <tsoome> it is in s1, which is enough if MBR does point to s1
[21:43:14] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 240 seconds)
[21:44:00] <tsoome> mbr reads sector 0 from s1, and Vbr from it will read sectors 1 to 273 (272 sectors) from s1
[21:44:13] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[21:44:27] <tsoome> but why there is no vbr in omnios image...
[21:45:35] <andyf> I'm sorry, I don't understand
[21:45:58] <andyf> After installboot .. s2, installboot -i shows the VBR block from s1 with an invalid signature
[21:47:19] <tsoome> where is the image download url?
[21:47:30] <tsoome> let me doublecheck:)
[21:47:33] <andyf> for the .usb-dd ?
[21:47:45] <tsoome> yes
[21:47:46] <andyf> https://downloads.omniosce.org/media/stable/omniosce-r151032.usb-dd
[21:52:47] <tsoome> yes, that image has MBR pointing to sector 0 in slice 2, but nothing there.
[21:53:41] <andyf> at least that explains the behaviour
[21:53:47] <andyf> thanks for your help with this
[21:54:53] <andyf> it looks like installboot s2 followed by installboot s0 installs the VBR blocks in the right palce
[21:54:55] <andyf> *place
[21:55:40] <andyf> I just sent v_a_b a test image which has had that done..
[21:56:20] <andyf> I should check what I'm doing against the OI image builder..
[21:57:48] <tsoome> there is a bug.
[22:02:26] *** bahamas10 <bahamas10!~dave@cpe-72-231-187-25.nycap.res.rr.com> has joined #illumos
[22:11:13] <andyf> Phew :)
[22:12:30] <v_a_b> The test image gives me messages, so that's a step ahead. "Can't find /boot/loader" and "Can't find /boot/zfsloader" :-)
[22:12:39] <v_a_b> tsoome andyf thanks for looking into this!
[22:13:12] <v_a_b> Once you have new media I'll be happy to test.
[22:21:18] <tsoome> anyhiw, the installboot still has bug and I’m checking out how…
[22:36:44] <jbk> i know there's been reports of issues w/ smartos usb images as well.. i wonder if they might be related
[22:41:47] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[22:48:41] <tsoome> I see where the problem is.
[22:50:14] *** cpk <cpk!~chris@185.172.87.163> has quit IRC (Ping timeout: 276 seconds)
[22:50:15] <tsoome> .o I think...
[22:54:08] <tsoome> ok, need to finish search tomorrow with fresh head...
[22:54:18] <andyf> np, and thanks!
[23:03:00] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has joined #illumos
[23:05:17] *** eki <eki!~eki@dsl-hkibng41-567327-143.dhcp.inet.fi> has quit IRC (Ping timeout: 240 seconds)
[23:05:30] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[23:07:08] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 276 seconds)
[23:07:23] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has quit IRC (Ping timeout: 252 seconds)
[23:08:01] *** neirac <neirac!~cneira@pc-184-104-160-190.cm.vtr.net> has joined #illumos
[23:18:52] <copec> I <3 all the lz4
[23:20:42] <AmyMalik> if i manage to hack zstd into zfs & if the lic is compat, who do i email with a unified diff?
[23:21:05] <andyf> Talk to AllanJude - he's already done it
[23:21:21] *** eki <eki!~eki@dsl-hkibng41-567327-143.dhcp.inet.fi> has joined #illumos
[23:22:32] <copec> andyf, any relation to axelf?
[23:22:32] <AmyMalik> oh, where do I send checks & money orders to help buy AllanJude some beef ribs
[23:23:16] <andyf> I think there's even a talk he did on it somewhere
[23:23:41] <AmyMalik> did his chg get rejected
[23:24:06] <andyf> no, as far as I know there are still a couple of things to resolve
[23:24:18] <AmyMalik> andyf: should i find a different comp algo to try to hack into zfs
[23:24:25] <andyf> https://www.youtube.com/watch?v=7N6lMi6g5UU
[23:24:57] <andyf> something around compressed ARC if I recall...
[23:24:59] <copec> https://github.com/zfsonlinux/zfs/pull/8044
[23:25:34] <copec> (ZoL is upstream for OpenZFS, and illumos merges from OpenZFS, or I could be screwing up the relationship)
[23:27:14] <andyf> Well, ZoL is about to be renamed to OpenZFS
[23:27:32] <andyf> and will become the repository for both ZoL and ZoF(reeBSD)
[23:27:37] <LeftWing> And eventually illumos!
[23:27:41] <LeftWing> :D
[23:27:59] <LeftWing> After the presumably considerable quantity of work to make that happen obviously
[23:28:36] <andyf> and ZoW(indows) and ZoM(acOS)
[23:29:02] <andyf> and eventually illumos
[23:30:04] <jbk> i.e. the plan is to turn 'ZoL' into a more portable/generic 'zfs' repo for all platforms
[23:35:53] <LeftWing> Right
[23:44:11] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has joined #illumos
[23:44:31] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has quit IRC (Quit: Leaving)
[23:50:07] <copec> There is no OpenZFS, only ZoL
[23:50:50] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[23:51:42] <jbk> heh.. a bit outside of san antonio on I-10 is an exit for 'Zuehl Rd'.. which always makes me think of ghostbusters
[23:52:12] <copec> hah
[23:55:22] *** andy_js_ <andy_js_!~andy@94.11.180.149> has joined #illumos
[23:56:56] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
top

   November 18, 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 | >