tazire

One thing I didn't see mentioned which is a big big plus for me about the array setup is that if you lose a drive beyond your parity drives you only lose the data on that one drive. You can just replace the drive and replace the data (assuming its not irreplicable) lost on only that drive. All the data on the other drives remains and is still usable. ZFS you will obviously lose everything should you lose that 3rd drive.... as unlikely as that maybe. In my case I never need the raw read speed of proper zfs (opposed to array zfs) for my media. And as long as there is not several streams from the same drive there is plenty of read speed available. And theoretically if I get multiple streams to different drives the cumulative array output speed can actually be very high (30 drives working at max output of 150-200MB/s say a conservative 100MB/s). Write speed is obviously limited to the speed of a single drive due to parity but this can be overcome with cache drives. But I dont put anything on my array that needs that raw write speed either. In the end I made multiple pools for data that I decided needed the speed and general benefits of zfs and then used the array for the easily replaceable media. I would say I am in a very different position to most in terms of the Hardware I'm running so these considerations might not be an option for everyone but definitely something to consider.

will do. Any idea why the MAC address isnt populating on the network settings? I assume this is just a completely different issue and not related?

Just looking for a bit of help with an FCP alert in relation to macvlan/bridging being active. So I made this post on general support about it. Essentially no matter what settings I have active it will give me the alert about macvlan/bridging being active. I have tried macvlan with bridging active which obviously gives the alert, then macvlan with bridging disabled which again gave the alert. macvlan with bridging disabled and making sure no containers have bridge set, and finally ipvlan active which will still give me the alert about macvlan and bridging being active. I made sure to reboot after each settings change. Everything seems to be working just fine with the exception of the MAC address not populating on the network settings page. Pic of this available on the above link along with diagnostics. any help would be appreciated.

Yea I did. I just rebooted again there now just to double check and still showing the error.

I've been doing some house keeping with my server recently. I didn't want to get into it while it was working well tbh. I've been using macvlan/bridging for as long as I can remember. I never had the crashing issue people reported but I did have some strange issues with network dropping out if I started to download through a VPN... might be a completely separate issue but just kind of assumed it was related. It never bothered me as it would return after 5 or 10 mins and everything would work as normal. So I had the FCP alert ignored. In the last few days I started to try address this alert. I followed the recommended settings from the docs. Firstly, as I have ubiquiti networking gear I tried to disable bridging and keep macvlan. FCP would continue to give me the alert saying that macvlan/bridging was active. I then went through each container to make sure they weren't set to bridging assuming that was causing it... still the alert from FCP. I then changed to ipvlan and turned back on bridging and still FCP persists in giving the alert. I will say that all containers have run perfectly in each setting. On a side note I have spotted in the network settings I noticed that the MAC address doesn't populate. Again this might have nothing to do with my issue. I'm on 6.12.8 but this issue has persisted since 6.12.6. Any help would be appreciated. server-diagnostics-20240216-1628.zip

Ah ok... Yea it seems that kasm is the issue... Thankfully its not vital for my needs and was just for playing with and being able to access blocked stuff in work. Thanks for your help m8.

How do I figure that out? Sorry never had to do it based off this information!

@Michael_P No i've no firefox container. Unless another container has it included. The only recent addition I've made was linuxservers kasm. But I dont leave any workspaces active. I have the settings such that it deletes the workspace after a certain amount of time inactive. But as a rule I will manually delete workspaces after I'm finished anyway.

Just need a bit of help identifying what could be causing this. I have attached my diagnostics. This is the 2nd time its happened. It doesnt happen every night but its the 2nd time over the last 6 months or so. The timing of the alert is usually between 4am and 5am. This is when appdata back up is running. This may or may not have anything to do with the issue. Usually by the time I wake up its resolved itself. Just curious if anyone can help. Cheers. EDIT*** I suspect the issue might be duplicacy. I think I changed the run time of appdata backup to be slightly later which possibly is causing an overlap in those backups. ie appdata backup is still running when duplicacy is starting its backup to B2. I'll try change the times to make sure this overlap is gone to test it. But if anyone can spot from the logs what the issue is that would be great. server-diagnostics-20240209-0904.zip

Very nice updates. I'm using that cooler and 2 of the Asus hyper cards as well. I've been half looking at an a4000 myself. Currently using a p4000. But I'm also tempted to hang on a while and get an ada generation a4000 for av1 support when the prices drop a bit. The p4000 is plenty for the time being anyway. I added a chenbro 24 bay 2u chassis for ssd's and moved everything to my spare room and put it in a lack rack! Eventually I'll move it out to the shed but I need to insulate and run power and networking before I can do that.

After a recent update I started to get the following error... The info on the link doesn't really address the issue. It mentions ocs-provider rather than ocm it mentions in the error. Very little info on google about the issue. I use NPM as my RP, not sure if that's the issue. Any help would be appreciated. Again this was never an issue until one of the recent updates.

The particular expander allows for up to 3 connections. Would this make any difference?

Curious if anyone knows anything about an LSI SAS3x36R expander? I have been looking at a 2U 24bay chassis that uses this expander with the intention of using it as a JBOD to connect SSD's to my server. I'm just trying to make sure that it will have the throughput to use SSD's. Fairly sure it should be absolutely fine during normal use but in a worst case if I'm hitting all 24 drives at once what's the theoretical throughput? EDIT* I would use a 9500-16e HBA to get 3 connections and PCIE4 link to HBA.

Thanks for the response. Just as you responded I had rebooted the server and the drive is back to normal operation and I can add it as a cache pool now. cheers.

Weird behaviour on an nvme drive. So I did have proxmox installed on this drive but I now want to clear the drive and use it as a cache drive on my test server. The drive appears twice on UD both at 8.59GB (512GB drive). When I try to clear or format the drive it just fails. Is there anything I can do to force clear/format the drive to get it back working as it should? I have a feeling this might just be whatever filesystem proxmox has on the drive causing the issue. Probably nothing to do with UD as such but any help would be great. tower-diagnostics-20230822-2339.zip