Switch to DuckDuckGo Search
   January 6, 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
[00:12:52] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 268 seconds)
[00:15:51] *** insomnia <insomnia!~insomnia@shadowcat/actuallyamemberof/lollipopguild.insomnia> has joined #illumos
[00:18:48] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[00:29:40] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 265 seconds)
[00:30:54] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[00:35:28] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 265 seconds)
[00:44:07] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[00:51:09] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Read error: Connection reset by peer)
[00:58:46] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[01:23:49] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Read error: Connection reset by peer)
[01:31:03] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[02:02:58] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-bvqxmwrzndkpjrfu> has quit IRC (Read error: Connection reset by peer)
[02:03:26] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-aktndwqlkbybmykz> has joined #illumos
[02:54:27] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-aktndwqlkbybmykz> has quit IRC (Read error: Connection reset by peer)
[02:59:32] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-spamiojmbyyvtixj> has joined #illumos
[03:03:18] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-spamiojmbyyvtixj> has quit IRC (Read error: Connection reset by peer)
[03:04:32] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-lumdeuhqqwotahpv> has joined #illumos
[03:31:15] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[03:34:47] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Read error: Connection reset by peer)
[03:51:47] *** X-Scale` <X-Scale`!~ARM@16.42.158.5.rev.vodafone.pt> has joined #illumos
[03:53:01] *** X-Scale <X-Scale!~ARM@46.50.7.110> has quit IRC (Ping timeout: 268 seconds)
[03:53:01] *** X-Scale` is now known as X-Scale
[04:07:10] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-lumdeuhqqwotahpv> has quit IRC (Read error: Connection reset by peer)
[04:07:19] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-clgsvvbmkgyjpddw> has joined #illumos
[04:12:44] *** X-Scale <X-Scale!~ARM@16.42.158.5.rev.vodafone.pt> has quit IRC (Ping timeout: 258 seconds)
[04:14:24] *** X-Scale` <X-Scale`!~ARM@31.22.203.158> has joined #illumos
[04:14:39] *** X-Scale` is now known as X-Scale
[04:16:56] *** tsun|z is now known as tsundoku
[05:35:51] *** BOKALDO <BOKALDO!~BOKALDO@81.198.17.49> has joined #illumos
[05:39:08] <richlowe> alloca taking a uint seems totally bogus?
[05:41:17] <richlowe> but this whole thing is super confusing
[05:53:44] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-clgsvvbmkgyjpddw> has quit IRC (Read error: Connection reset by peer)
[05:54:25] *** heroux <heroux!sandroco@gateway/shell/insomnia247/x-vxpubizgfvaclowm> has joined #illumos
[05:54:46] <richlowe> looks like the builtin used to take a uint, but now takes a size_t but warns if the argument doesn't fit a uint?
[05:54:58] <richlowe> also, if you try to alloca something that doesn't fit in a uint, it should just catch on fire to spite you.
[05:54:59] <richlowe> you deserve it
[06:09:00] *** Kruppt <Kruppt!~Kruppt@50.111.56.62> has quit IRC (Quit: Leaving)
[06:51:10] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Quit: amrfrsh)
[07:23:45] <tsoome> well, it seems they are making noise around it to get people to stop using alloca
[07:29:07] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has joined #illumos
[07:44:54] *** pwinder <pwinder!~pwinder@86.4.7.64> has joined #illumos
[07:54:12] *** tsoome <tsoome!~tsoome@80.235.52.148> has quit IRC (Quit: tsoome)
[08:15:23] *** kev009 <kev009!~kev009@ip72-222-200-117.ph.ph.cox.net> has quit IRC (Remote host closed the connection)
[08:18:05] *** kev009 <kev009!~kev009@ip72-222-200-117.ph.ph.cox.net> has joined #illumos
[08:40:16] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[08:50:42] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has quit IRC (Quit: ^C)
[08:52:51] <tsoome> “Fundamentally, the problem seems to be that Javascript programmers should not be allowed to design APIs for C code.”
[08:54:12] <sensille> oh, you can do chaining in C
[09:19:52] <sjorge> bahamat: poke, for LDAP mirrormode ldifs :)
[09:21:34] <v_a_b> What is "mirrormode"? Something like replication?
[10:09:52] *** pwinder <pwinder!~pwinder@86.4.7.64> has quit IRC (Quit: Leaving)
[10:14:43] *** pwinder <pwinder!~pwinder@86.4.7.64> has joined #illumos
[10:16:37] *** man_u <man_u!~manu@manu2.gandi.net> has joined #illumos
[10:27:12] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has quit IRC (Ping timeout: 260 seconds)
[10:31:01] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has joined #illumos
[10:42:51] <sjorge> Yeah
[10:43:01] <sjorge> One active write and multiple readers
[10:43:18] <sjorge> But I'd just chuck vrrp in front so it is HA without the gains in read performance
[10:48:09] <v_a_b> Ah OK. About a year ago I set up a pair of OpenLDAP servers on Solaris 11.4 using multi-master replication.
[10:52:03] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has quit IRC (Quit: amrfrsh)
[10:55:41] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has joined #illumos
[10:57:36] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has quit IRC (Client Quit)
[11:04:42] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has joined #illumos
[11:06:42] *** amrfrsh <amrfrsh!~Thunderbi@185.212.171.68> has quit IRC (Client Quit)
[11:11:12] *** elegast <elegast!~elegast@83-161-163-206.mobile.xs4all.nl> has joined #illumos
[11:19:52] <Agnar> v_a_b: does OpenLDAP recover correctly from multi-master failure? or will it still have inconsitency?
[11:26:53] <tsoome> what is the size estimate to justify multimaster setup?
[11:27:16] <sjorge> I just want HA my self, hench mirrormode seems easier
[11:57:14] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has joined #illumos
[12:19:21] *** leah2 <leah2!~leah@vuxu.org> has quit IRC (Ping timeout: 252 seconds)
[12:34:52] *** jollyd <jollyd!~alarcher@2a01:cb1d:74b:4700:3f58:765c:cb5:aac5> has quit IRC (Quit: Leaving.)
[12:41:40] *** leah2 <leah2!~leah@vuxu.org> has joined #illumos
[12:43:10] *** leah2 <leah2!~leah@vuxu.org> has quit IRC (Client Quit)
[12:43:51] *** leah2 <leah2!~leah@vuxu.org> has joined #illumos
[12:48:13] <v_a_b> Agnar Good question! I stopped one server once and after restart, synch resumed. I really should do more tests. Also, there is no monitoring. So far, the customer hasn't complained. :-)
[12:50:01] <v_a_b> tsoome About 10 GZ, 30-40 NGZ, some ILOMs, some VMware guests. It's more a question of redundancy. Load balancing could be done via DNS, but we just set up one client using profile #1, and the next using profile #2. Both profiles are identical except for the "preferred servers" sequence.
[12:50:54] <tsoome> ok
[13:13:30] *** BOKALDO <BOKALDO!~BOKALDO@81.198.17.49> has quit IRC (Quit: Leaving)
[13:14:28] <tsoome> am… git pbchk is getting blown up:
[13:14:29] <tsoome> beastie -> 165.225.151.82 HTTP GET /issues/1317.xml HTTP/1.1
[13:14:30] <tsoome> 165.225.151.82 -> beastie HTTP HTTP/1.1 302 Moved Temporarily
[13:21:18] <wilbury> as usual.
[13:27:46] <Agnar> v_a_b: that's why I chose OpenDS/OpenDJ, it has a replication-transaction log and can survive and resync multi-master failures
[13:36:03] <wilbury> yeah, i used opendj for years, 3-node circular multimaster, stable AF
[13:42:22] *** neirac <neirac!~neirac@pc-184-104-160-190.cm.vtr.net> has quit IRC (Ping timeout: 258 seconds)
[13:46:04] <Agnar> same here, 3 OpenDJs
[13:51:15] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has joined #illumos
[13:53:23] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Ping timeout: 260 seconds)
[13:54:58] <andyf> tsoome - because pbchk uses http:// do you think?
[13:55:49] <andyf> the redirect should be handled by python requests., but the XML download seems to be prompting for login details
[13:57:49] <andyf> http://illumos.org/issues/1234.xml -> https version -> https://www.illumos.org/issues/1234.xml
[13:57:53] <andyf> the latter prompts for authentication
[13:58:02] <andyf> FWIW, it worked this morning when I prepared an RTI
[14:04:05] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:07:05] *** BOKALDO <BOKALDO!~BOKALDO@81.198.17.49> has joined #illumos
[14:35:14] *** hemi770_ <hemi770_!~hemi666@unaffiliated/hemi770> has joined #illumos
[14:36:28] *** hemi770 <hemi770!~hemi666@unaffiliated/hemi770> has quit IRC (Ping timeout: 265 seconds)
[14:39:52] *** hemi770_ is now known as hemi770
[14:55:07] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[15:07:38] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has quit IRC (Ping timeout: 240 seconds)
[15:17:29] <v_a_b> Agnar Yes, I just migrated away from OpenDJ. I disliked the anti open source attitude of Forgerock, and it wasn't possible to get current versions of the software any more.
[15:17:58] <Agnar> v_a_b: https://github.com/OpenIdentityPlatform/OpenDJ/releases/tag/4.2.5
[15:18:00] <Agnar> :)
[15:18:08] <v_a_b> The product itself is rock-solid and well worth using in production when you want to pay for it.
[15:18:19] <Agnar> https://github.com/OpenIdentityPlatform/OpenDJ/releases
[15:18:21] <Agnar> sorry
[15:19:18] <v_a_b> Also, they explicity removed all support for Solaris. And I did not have the time and expertise to restore everything.
[15:19:48] *** leah2 <leah2!~leah@vuxu.org> has quit IRC (Ping timeout: 248 seconds)
[15:19:54] <Agnar> well, I run the latest on solaris with OpenJDK8
[15:20:37] <v_a_b> Interesting... the last time I looked there was no "latest" source available. There were people working on a fork but there wasn't really anything usable. That was a while ago...
[15:21:31] <Agnar> v_a_b: jepp, but suddenly I stumbled over this github repo and it made me very happy ;) besides that: I do not really need new features (yet?) and I never ran into a bug
[15:22:03] <v_a_b> Well, too late for me. :-(
[15:22:27] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has joined #illumos
[15:22:29] <v_a_b> Maybe I will play with it when I have done the other 10000 things I want to do...
[15:23:06] *** leah2 <leah2!~leah@vuxu.org> has joined #illumos
[15:38:33] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Remote host closed the connection)
[15:40:46] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[15:43:20] *** Kruppt <Kruppt!~Kruppt@50.111.56.62> has joined #illumos
[15:47:22] *** neirac <neirac!~neirac@pc-184-104-160-190.cm.vtr.net> has joined #illumos
[15:47:24] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[16:21:45] *** andy_js <andy_js!~andy@90.215.171.48> has joined #illumos
[16:26:23] *** andy_js <andy_js!~andy@90.215.171.48> has quit IRC (Client Quit)
[16:49:26] *** kev009_ <kev009_!~kev009@ip72-222-200-117.ph.ph.cox.net> has joined #illumos
[16:51:45] <elegast> I'm still slightly confused about how loader works with exposing files through bootfs; specifically with regards to paths:
[16:51:48] *** kev009 <kev009!~kev009@ip72-222-200-117.ph.ph.cox.net> has quit IRC (Ping timeout: 265 seconds)
[16:52:41] <elegast> my_module_type=file, my_module_name=/bin/my_module
[16:53:29] <elegast> however, it shows up only on /system/boot/my_module where i'm expecting it ALSO on /bin/my_module
[16:54:21] <elegast> is there something I'm missing?
[16:55:40] <jlevon> did you see some reference that led you to expect that
[16:56:12] <elegast> jlevon: when using grub, that is how bootfs worked, was it not?
[16:56:41] <jlevon> andyf: which fdinfo gerrit should I be reviewing now? or should I hold off a while?
[16:57:03] <jlevon> elegast: loader isn't grub though.
[16:57:43] <andyf> jlevon: I'm just about to push an update to address rmustacc's last few comments, and with the documentation updates
[16:57:50] <jlevon> ok great.
[16:57:52] <andyf> jlevon: well, after I do one last test build, so in an hour os so
[16:58:17] <andyf> It's still https://code.illumos.org/c/illumos-gate/+/262
[16:58:26] <jlevon> ok thanks. will look out
[16:58:53] <elegast> jlevon: ah, true. But I believe the whole point of bootfs getting implemented was to override driver conf files, without replacing files in the platform archive?
[17:00:48] <jlevon> the problem is that your source path has to match exactly
[17:01:19] <jlevon> hmm actually
[17:01:49] <elegast> ok so now the source file is /mnt/usbkey/bin/my_module, should it instead be /mnt/usbkey/boot/bin/my_module?
[17:02:04] <tsoome> /bin is symlink in actual system
[17:02:06] <elegast> If it was false, I would've expected the file to not be loaded at all
[17:04:15] <elegast> tsoome: ah, thanks, that detail had gone by me. I'll account for that, though, not sure if that'll fix the issue
[17:06:34] <tsoome> I have only used to override /etc/system file myself
[17:07:09] <elegast> tsoome: I will test with placing a file in /etc/xxx first.
[17:17:59] <elegast> /mnt/usbkey/etc/xxx, and in loader.conf: xxx_name=/etc/xxx, exact same behaviour ;(
[17:18:56] <tsoome> elegast: the MODULE_name does set where from to load the file, now if you want to have it “mounted” in different location, provide MODULE_flags=“name=/path/file”
[17:21:25] <jlevon> TIL
[17:23:23] <tsoome> elegast: see https://paste.ec/paste/JqQ+3mSL#ge8mC1xNDamAacQatPDP5QrxMJ2iwoeNutgSQHW2SzJ
[17:23:26] <elegast> tsoome: ah ... I feel slightly stupid. jlevon: I used your blog as a reference, but the MODULE_flags isn't mentioned, that would be a great addition
[17:23:56] <jlevon> yeah, I need to update it now.
[17:24:02] <tsoome> if you care, please add example into loader.conf manual:)
[17:25:11] <jlevon> sure
[17:25:32] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[17:25:59] <elegast> tsoome, jlevon: many thanks to the both of you
[17:26:15] <tsoome> np
[17:51:02] <jlevon> elegast: I updated https://blog.movementarian.org/2019/02/modifying-boot-files-with-smartos-under.html
[17:53:59] <rmustacc> so nothing about bootfs causes files to be replaced, fwiw.
[17:54:11] <rmustacc> It's just that the kernel search path has bootfs first.
[17:55:42] <jlevon> the /bin thing isn't really relevant here yeah. it's just that the setting above meant it defautled to /system/boot/my_module
[17:55:43] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Quit: man_u)
[18:11:05] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has quit IRC (Read error: Connection reset by peer)
[18:12:59] <elegast> rmustacc: so even though after system boot, /etc/system contents would be the one from the PI, but during boot, while driver options are set/get, it reads the contents from bootfs/etc/system? Not quite sure I understand the implications/goals here
[18:13:10] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has joined #illumos
[18:14:59] <jlevon> elegast: from /system/boot/etc/system yes
[18:15:14] <jlevon> elegast: it finds that first so ignores /etc/system
[18:15:46] <jlevon> if you wanted to arbitrarily modify stuff in the root fs, you could add an smf service to loopback mount from /system/boot/
[18:17:59] *** pwinder <pwinder!~pwinder@86.4.7.64> has quit IRC (Quit: This computer has gone to sleep)
[18:56:55] *** nde <nde!~nde@96.79.76.105> has joined #illumos
[19:00:57] *** nde <nde!~nde@96.79.76.105> has quit IRC (Read error: Connection reset by peer)
[19:10:11] <gitomat> [illumos-gate] 12078 loader: rewrite zfs vdev initialization -- Toomas Soome <tsoome at me dot com>
[19:10:11] *** pwinder <pwinder!~pwinder@86.4.7.64> has joined #illumos
[19:14:28] *** nde <nde!~nde@96.79.76.105> has joined #illumos
[19:23:33] *** pwinder <pwinder!~pwinder@86.4.7.64> has quit IRC (Quit: This computer has gone to sleep)
[19:27:12] *** nde <nde!~nde@96.79.76.105> has quit IRC (Remote host closed the connection)
[19:40:06] *** nde <nde!~nde@96.79.76.105> has joined #illumos
[19:44:03] *** pwinder <pwinder!~pwinder@86.4.7.64> has joined #illumos
[19:50:44] *** nde <nde!~nde@96.79.76.105> has quit IRC (Remote host closed the connection)
[19:53:12] <bahamat> sjorge: ping
[19:54:34] *** pwinder <pwinder!~pwinder@86.4.7.64> has quit IRC (Quit: This computer has gone to sleep)
[20:26:18] *** zsj <zsj!~zsj@3EC95F11.catv.pool.telekom.hu> has quit IRC (Quit: leaving)
[20:37:35] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has joined #illumos
[20:45:09] <gitomat> [illumos-gate] 12143 scan code should check the return value of zfs_btree_first -- Paul Dagnelie <pcd at delphix dot com>
[20:55:44] *** alanc <alanc!~alanc@inet-hqmc01-o.oracle.com> has joined #illumos
[20:58:56] *** neirac <neirac!~neirac@pc-184-104-160-190.cm.vtr.net> has quit IRC (Read error: Connection reset by peer)
[20:59:15] *** neirac <neirac!~neirac@pc-184-104-160-190.cm.vtr.net> has joined #illumos
[21:06:23] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has quit IRC (Read error: Connection reset by peer)
[21:08:35] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has joined #illumos
[21:13:20] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has quit IRC (Ping timeout: 268 seconds)
[21:21:30] *** BOKALDO <BOKALDO!~BOKALDO@81.198.17.49> has quit IRC (Quit: Leaving)
[21:29:43] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has joined #illumos
[21:35:21] <gitomat> [illumos-gate] 12148 libc: _rtbootld.o needs to be built with -m32 -- Toomas Soome <tsoome at me dot com>
[21:39:37] *** elegast <elegast!~elegast@83-161-163-206.mobile.xs4all.nl> has quit IRC (Ping timeout: 258 seconds)
[21:43:55] <sjorge> bahamat pong, but I assume the privmsg ws the reason?
[21:44:06] <bahamat> Yeah.
[21:45:16] <bahamat> Let me know if you have any questions. The way we do it, we have a mirrormode pair that's the primary copy of the database, then we have replica-only nodes spread across each region.
[21:45:45] <bahamat> I've never tried it, but I believe that a replica can be promoted to a mirrormode primary by adjusting the configs appropriately.
[21:46:46] <bahamat> Where replication occurs using direct connections we use streaming (i.e., refreshAndPersist), so the primary can push updates to the replicas in real time
[21:47:20] <bahamat> But where we have to traverse a NAT (which almost always means suffering from session timeouts) we use refreshOnly, and set a polling interval.
[21:47:53] <bahamat> So using this model, you can scale as large as you need to.
[21:48:58] <bahamat> We don't use vrrp or carp, but it should work great with that, and you can probably use haproxy to load balance across a cluster in a region.
[21:50:04] <gitomat> [illumos-gate] 12149 mixed up 32 and 64 bit build -- Toomas Soome <tsoome at me dot com>
[21:53:54] <sjorge> I think refreshAndPersist on both should work right, as connections will only go to one?
[21:54:12] <sjorge> Hopefully I'll have some time next weekend :)
[21:55:21] <bahamat> As long as you have a direct connection, refreshAndPersist is best.
[21:58:41] <sjorge> Yeah it's 2 CN's on the same switch
[21:58:48] <sjorge> And it will rarely be written too
[22:01:26] *** AllanJude <AllanJude!~allan@freebsd/developer/AllanJude> has quit IRC (Remote host closed the connection)
[22:12:57] *** AllanJude <AllanJude!~allan@allanjude.bsdbasement.com> has joined #illumos
[22:14:12] <bahamat> Yeah, mirrormode with refreshAndPersist is what you want.
[22:14:59] <bahamat> I've been able to (several times) wipe out one of the mirrormode instances, and rebuild it simply by starting up slapd with the proper configuration.
[22:16:33] <bahamat> For ultimate safety, a streaming replica in another region could be used to slapcat/slapadd for bootstrapping a mirrormode cluster.
[22:18:06] <bahamat> And if you haven't seen it yet, this is our directory builder: https://github.com/bahamat/ldap_kit
[22:18:34] <bahamat> We use that to bootstrap a new dit.
[22:39:35] <jbk> LeftWing: should a duplicate ticket be closed or resolved (after linking to the original) and is there anything special that needs to happen so any automation is happy?
[22:39:53] <LeftWing> I think Closed is fine
[22:39:56] <jbk> ok
[22:40:14] <LeftWing> Make it a related issue too?
[22:40:17] <LeftWing> I think
[22:40:33] <jbk> yeah, i added a 'duplicate' in there
[23:10:05] *** hawk <hawk!~hawk@d.qw.se> has quit IRC (Quit: WeeChat 2.4)
[23:12:00] *** hawk <hawk!~hawk@d.qw.se> has joined #illumos
[23:12:19] *** nde <nde!~nde@c-98-220-146-104.hsd1.in.comcast.net> has joined #illumos
[23:13:39] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has quit IRC (Quit: Leaving)
[23:14:40] *** nde <nde!~nde@c-98-220-146-104.hsd1.in.comcast.net> has quit IRC (Read error: Connection reset by peer)
[23:19:08] <gitomat> [illumos-gate] 12141 libbe cannot handle vfstab updates in NGZ -- Andy Fiddaman <omnios at citrus-it dot co.uk>
top

   January 6, 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