!VhzbGHamdfMiGxpXyg:robins.wtf

NixOS LXC

33 Members
lxc, lxd, incus discussions related to NixOS14 Servers

Load older messages


SenderMessageTime
10 Mar 2024
@hexa:lossy.networkhexaI do19:02:58
@hexa:lossy.networkhexawas just trying to understand the alternatives19:03:12
@adam:robins.wtfadamcstephenshttps://github.com/lxc/incus/issues/56419:03:48
@adam:robins.wtfadamcstephenshttps://github.com/lxc/incus/issues/34419:04:05
@hexa:lossy.networkhexaglusterfs is some weird distributed filesystem 😄 19:05:09
@hexa:lossy.networkhexalinstor is iirc the firm behind drbd19:05:19
@adam:robins.wtfadamcstephensyep19:05:26
@hexa:lossy.networkhexaor linbit is19:05:31
@hexa:lossy.networkhexaI don't want to go back to drbd 😄 19:05:45
@hexa:lossy.networkhexaso what is linstor19:05:53
@hexa:lossy.networkhexa

LINSTOR has multiple uses, but it is typically used to manage a large number of DRBD replicated storage resources

19:06:53
@adam:robins.wtfadamcstephensi've never used drbd 🤷‍♂️19:07:16
@hexa:lossy.networkhexaway back, when we had two node hypervisor clusters over 10y ago19:07:30
@adam:robins.wtfadamcstephensyeah, gluster is also an old filesystem. but it's dying 19:10:00
@adam:robins.wtfadamcstephensi think the real problem is clustered/distributed filesystems are hard and RH is probably the only one making any serious money with them19:12:24
@adam:robins.wtfadamcstephens(in the open source space)19:12:40
@hexa:lossy.networkhexaso what clusterlvm does is make sure that in a clustered lvm setup that only one node can write metadata19:16:09
@hexa:lossy.networkhexathey share the same block device, which can be attached by any means, but most likely iscsi19:16:27
@hexa:lossy.networkhexathat only makes sense if you have a chonky multihead storage cluster19:16:41
@hexa:lossy.networkhexaI'd rather do hyperconverged setups, and that is where ceph shines19:17:06
@hexa:lossy.networkhexathe main benefit is that I can rely on commodity hardware19:18:57
@hexa:lossy.networkhexaanyway, one weird thing I imagined is zfs send/recv style migrations19:19:26
@hexa:lossy.networkhexaI think proxmox could do that19:19:36
@hexa:lossy.networkhexaand it might be appealing since zfs has alot less moving parts than ceph19:19:47
@adam:robins.wtfadamcstephensat least for non-clustered incus, i think it does use zfs send/recv for migrations19:34:21
@adam:robins.wtfadamcstephensit may for clustered too when it's not a shared volume19:34:42
@adam:robins.wtfadamcstephens * it may for clustered too when it's not a shared pool19:34:45
@cmspam:matrix.orgcmspam joined the room.19:34:51
@cmspam:matrix.orgcmspamHello, everyone.19:40:11
@mkg20001:mkg20001.iomkg20001
In reply to @adam:robins.wtf
I’m also still hoping to deprecate lxd. I’ve just been waiting for incus LTS 
i'll make it seperate packages then
19:42:03

Show newer messages


Back to Room ListRoom Version: 10