Hey guys, 

 

Tldr: Internet not working on unRAID server

 

I'm having some network issues with my unRAID server. This started when my pihole stopped working. I looked into the logs and I think pihole couldn't reach the update server or something(this was like a few weeks ago and I have since deleted the pihole container). After that I started getting a 'cant communicate with GitHub message. After a bit of investigation, I noticed the internet cuts out just a few minutes after turning on the server. It does come back but it's after a few hours/days and will periodically cut out sgain. To test this theory, I rebooted multiple times and tried to ping google.com. I was able to ping it after rebooting, but I tried again after a few minutes and I got a 'name or service not known' error. One time I tried pinging Google and it worked but I got like 50 percent packet loss.

 

I'm also seeing other errors like my docker containers versions being set to 'not available'. I also can't access any of my reverse proxy stuff nor my pihole. However, I am able to access things like Plex and krusader locally from most machines but sonarr can't fetch new episodes and Plex can't fetch metadata.

 

After searching online, I've tried to change the dns severs in the unRAID network settings to Google servers, checking physical connections and a few other things but no luck. Also noticed that my router didn't have my unraid server's ip address under the dhcp sever (not sure if it was there before so I manually created an entry for it. I'm attaching diagnostics of my server. I've tried looking through this myself but I'm not sure what I'm looking for/at. Thanks in advance!

 

Ps. Sorry for the long post, trying to provide as much details as possible so that hopefully something clicks with someone and they know how to fix this.

swagserver-diagnostics-20211115-1228.zip

@trurl any ideas or if you know how we can get some viability on this? Thanks!

Not sure about your network issues, but your cache pool has checksum errors.

 

Have you done memtest?

@trurl thanks for your response. I've been running memtest for like 6 hours and no errors so far. I did recently add 2 drives to the cache pool so thay could be a reason.

Also, would you mind explaining how you look though the diagnostics/what you look for?

Those checksum errors are obvious in syslog. Memory is a possible cause of bad btrfs data.

 

Whether or not memory is bad, those need to be fixed. Let's see what @JorgeB thinks.

 

 

Run a scrub on the pool, if all errors are correctable then run a balance to raid1, pool has dual data profiles, you should get a notification about that, if they are enable.

 

Also see here for better pool monitoring.