| ham5urg | Does the linux-modules-x.xx.x-xx-generic package exist in some form? | 00:01 |
|---|---|---|
| * debdog cannot find any | 00:08 | |
| debdog | except virtual ones | 00:09 |
| golinux | https://pkginfo.devuan.org/cgi-bin/policy-query.html is your friend | 00:33 |
| ted-ious | Is not being able to apt update daedalus-security and daedalus-backports expected? | 06:16 |
| ted-ious | Or did I break something? | 06:16 |
| ted-ious | I think I broke something. | 06:17 |
| micdud | daedalus is testing so it wont have security nor backports | 06:32 |
| micdud | once it goes stable then yes | 06:32 |
| ted-ious | Oh ok. | 06:37 |
| ted-ious | I don't remember it throwing these errors when I did the dist-upgrade. | 06:38 |
| ted-ious | I also get errors about not being able to create /tmp/apt.conf.abcdefg for passing config to apt-key. | 06:39 |
| onefang | Daedalus is about to move to stable soonish or laterish, THEN it'll have daedalus-security and daedalus-backports. | 06:50 |
| rrq | ted-ious: daedalus-security and daedalus-backports are set up; as are even daedalus-updates and daedalus-proposed-updates, although they are all "empty" atm, until daedalus becomes the table release. | 07:00 |
| rrq | which error do you get? | 07:00 |
| onefang | Ah rrq is correct. B-) | 07:01 |
| rrq | right now, "daedalus" is frozen in the state of having been the (rolling) testing release, and about to be the "current stable" release. Technically it's still "testing" but as such daviating from Debian's testing which has moved on a fair but from the time of daedalus freeze | 07:06 |
| rrq | and realistically it is already the stable release although not yet aliased as that. | 07:09 |
| rrq | if we find us an AI to help us over the last adminsitrative hurdles it'll all be done in no time | 07:11 |
| al1r4d | i usually use devuan unstable and rarely to get problem | 08:48 |
| brocashelm | i am able to get upgrades for daedalus-proposed-updates and daedalus-updates, although that's due to debian bookworm versions of those that are being merged (so maybe the non-systemd stuff is yet to be pushed) | 09:16 |
| brocashelm | daedalus is technically *stable*; it's just not officially released yet (the isos are needing more testing atm) | 09:16 |
| systemdlete | I wonder if I have run into some more "damage" caused by u-know-what. I am trying to configure a PC with 2 operating systems and 2 (what I would like to be 2 matching) lvm configurations. This way, only the VG associated with the booted OS will be auto-activated. Trouble is, I can't seem to figure out how to get an OS to automatically activate ONLY the VG for that OS. | 17:28 |
| systemdlete | I tried to set the auto-activation on each VG, and then set the auto-activate-volume-list option in lvm.conf to just that VG. It didn't work; all LVs were activated at boot. I even tried the "skip" option but I can't seem to get that to work either. | 17:30 |
| systemdlete | I realize this is probably a bit esoteric, but it would be nice, just to avoid accidentally "touching" an LV that does not belong to the booted OS. | 17:31 |
| * systemdlete is trying to keep systemdlete from making silly mistakes... | 17:31 | |
| systemdlete | the man page for lvm activation indicates that VGs are activated by events in the boot process, and SEEMS to allude to systemd being the instigator. So I am thinking that eliminating it from devuan has caused a minor difference in LVM behavior. But that might be stretching logic here, idk. | 17:34 |
| systemdlete | Btw, this is all chimaera on physical HW, no VMs involved at this point. | 17:35 |
| mason | systemdlete: Are you trying to get things ignore on one side, the other, or both? | 17:35 |
| systemdlete | mason: thanks. Let me label some things for reference, ok? | 17:36 |
| mason | systemdlete: Well. Let me just throw this at you for now: https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/logical_volume_manager_administration/lvm_filters | 17:36 |
| systemdlete | Let's call the two OS's "live" and "test" | 17:36 |
| mason | Probably you want to set up specific filters on each side to ignore the other side. | 17:36 |
| systemdlete | And let's say I have two VGs, "live" and "test" | 17:36 |
| systemdlete | I would like to boot each OS so that it only activates the VG with the same name. | 17:37 |
| mason | Yeah. Look at filters. | 17:37 |
| systemdlete | mason: There is a lvm.conf file also. | 17:37 |
| mason | I'm not expert with them but I think they're what you want, and you'd implement them in that lvm.conf | 17:37 |
| systemdlete | Supposedly it can do what I want (I think). | 17:37 |
| systemdlete | (we might be talking about the same thing, mason!) | 17:38 |
| mason | Example from the docs: The following filter adds just partition 8 on the first IDE drive and removes all other block devices: filter = [ "a|^/dev/hda8$|", "r/.*/" ] | 17:38 |
| mason | I suspect we are. | 17:38 |
| mason | don't forget to update initramfs on each side once you've set the filters, as I believe that's copied in and used. | 17:39 |
| mason | Yeah, lvm.conf is copied in and the copy inside your initramfs is probably going to matter a lot for this. | 17:39 |
| systemdlete | I'm not useing "filter = ..."; I am using auto_activation_volume_list | 17:40 |
| systemdlete | oh! | 17:40 |
| systemdlete | crud! | 17:40 |
| systemdlete | thank you... | 17:40 |
| systemdlete | you know, I was thinking about that, momentarily, but didn't follow up on that. | 17:40 |
| systemdlete | I think you may have just nailed it. | 17:40 |
| mason | It ought to work for you, but make a noise if it presents issues. | 17:40 |
| systemdlete | let me try re-gen'g the initramfs | 17:41 |
| systemdlete | mason: thanks again | 17:41 |
| mason | sure thing | 17:41 |
| * systemdlete continues reading "101 Ways to Bite Yourself on Your Own A$$" | 17:42 | |
| mason | Technology amplies our capabilities in this regard boundlessly. | 17:42 |
| mason | amplifies | 17:42 |
| * systemdlete nods | 17:42 | |
| systemdlete | There are at least 3 man pages dealing with various aspects of lvm config and activation... | 17:43 |
| mason | systemdlete: Yeah. That's why stuff with examples is useful. That Red Hat documentation should be relevant. | 17:52 |
| mason | I'm no longer in a position to pop knowledgebase articles public, but for now at least the formal documentation is still freely licensed and not paywalled. | 17:53 |
| systemdlete | Actually, I think I might have it now. I set the VG I don't want activated to "skip"-- I think I just need to regen the initramfs so early boot will see the config I want for each OS | 17:54 |
| systemdlete | I won't know for certain until I install the 2nd OS. But I think this might work. If not, I'll come back here crying for help... | 17:55 |
| mason | Unfamiliar with "skip". | 17:55 |
| mason | kk | 17:55 |
| systemdlete | skip allows the volume group (or individual LV) to be skipped by pvscan | 17:56 |
| systemdlete | but, really, all of this gets tricky with so many options. | 17:56 |
| systemdlete | Utlimately, it is probably initramfs that will save my bacon here. | 17:56 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!