Search results

Do not use LACP for corosync network: instead use two corosync links over a simple nic each and let Corosync detect link failures. I doubt so. Never had an issue with corosync related to kernel version. Impossible to ascertain with the commands you posted. If this ever happens, you will need...

Well, my brain kept just the "fortnight" part and forgot the "Sunday" one. Still gave you a pointer and a link to the documentation that could have followed too...

Remember that LXC logs are also seen from the host itself. Your container is setting it's own nic (eth0) in promiscuous mode (i.e. will be able to see all traffic in that L2 segment), not the host's nic.

Using more than one OSD per drive allowed to workaround internal processing limits in the OSD services themselves. Your Intel P4610 aren't the fastest drives ever, so chances are that you won't hit those limits. With recent versions of Ceph this adds very little real world benefit, which is...

By "px" you mean "pve"? Which issue exactly are you refering to? With PBS you backup the VM and can recover both the whole VM and files/directories inside the VM's OS, though you can't recover files/directories directly inside the VM: you have to download to an admin PC and upload them back to...

Something like this will work: *-*-*/15 03:00 "make a backup any year, any month, on days divisible by 15 at 3:00 AM" You could also do: *-*-1,15 03:00 "make a backup any year, any month, on days 1 and 15 at 3:00 AM" Details here.

Yes you can as long as you only one host. None that I'm aware off. It's essentially a LUN connected via FC from that host to the storage. Yes it is, as long as only one server accesses that LUN. Yes you do: storage replication requires ZFS. ZFS does not support any kind of SAN/RAID and...

Now we are talking! It's doable, but will require some steps and has some restrictions. The idea is to restrict pool zpool1 to hosts pve2,3 and 4 and rename zpool1 in pve1 to filepool. Then add disk(s) to pve2 and create a filepool there. Now you have filepool in two machines (1,2) and zpool1...

IMHO all are misuses of an unprivileged LXC: it exists to isolate the LXC from the host as much a possible and that's why it is hard to "un-isolate" the LXC from the host. As simple as that. In this very case of a PBS, an unprivileged LXC adds nothing but headaches (but you already noticed...

Try to get PBS in a completely different machine to ease the recovery when the PVE host(s) fail, at least the one on the "main" site. That would also completely workaround the permissions issue. Remember that you can also use a Virtual Machine and use USB or full disk pass through to get the VM...

It's impossible to give accurate instructions/recommendations without accurate questions/configurations. Please, post the exact config of the hosts regarding disk/storage so we can help you out instead of pulling a crystal ball to guess your settings :)

IIUC, you have one host with some big pool were you host your fileserver VM. Let's call this pool "filepool" and shows as storage "filepool" in PVE, restricted to the one host that runs the fileserver VM. You would just need to add disk(s) to another host, configure a ZFS with pool name...

IMHO, using an unprivileged LXC for PBS doesn't make sense. The isolation that provides an unprivileged container is of no use in this use case. I mean, I don't expect PBS processes to misbehave or become malware that may try to get out the LXC via the host's kernel and try to break havoc in the...

This effectively renders the cluster useless: once you lose any OSD, there will be no I/O on the PGs stored in that OSD until they get recovered from the single copy still in the cluster. You should always use at least size=3, min_size=2 unless you can tolerate such downtime. Not to mention the...

AFAIK, there's no way to do that in a sensible way. The main reason for it is that all namespaces in the same datastore share the same set of chunks. That means that data is deduplicated among all namespaces of a datastore. If there was a way to set quotas to a namespace, which one would you...

Is this expected to be released with PVE9? Need to implement a solution for a use case were backup selection using tag would fit perfectly. Thanks!

If you install qemu-guest-agent on the VM the backup should be as consistent as with stop mode, unless you have some strange application there. Using two namespaces will require PVE to use two PBS storage, thus losing dirty bitmap when doing backups to the other, although the stop mode backup...

Given what's mentioned a few posts above [1], it could be released with PVE9 that will be based on Debian Trixie. Given that Trixie has not reached full freeze yet [2] but expecting that may happen soon, we could expect a PVE9 release around Q3 this year (it think I've seen some info posted by...

If you just want NFS, then maybe for some use cases multipath is better. If you want to cover other use usages, like normal VM traffic, were multipath doesn't apply/exists, bonding is an easy way to aggregate links.

Create a bonding at the PVE level using two physical nics. Then add a bridge over that bond. Your host will deal with load balancing and redundancy of the network connection. Which bonding mode to use depends on the switches and their features (i.e. you need stacking or MLAG to use LACP 802.3ad...