February 27, 20215 yr Hello everyone, long time lurker, first time poster. I can usually troubleshoot my way out of most issues but feeling stumped... My setup is pretty basic: Unraid Plus, 9 disks + single parity, 1-4GB in size, SSD for cache, SSD for unassigned devices. All leftover commodity hardware, humming along nicely for a couple of years with just 1 disk change. Used as a media server, *arr stuff, couple of other small dockers, 1 VM for random lab stuff. Network is dirt simple, 192.168.3.0/24 subnet, ISP router issues IPs to regular end clients for 3.20+ and everything below is static (unraid, two pi holes, printer, etc). It's been humming along perfectly and then two days ago I had some spare cycles and decided to do some work on it. * I upgraded RAM from 8GB to 24GB (pre-tested with memtest for a 24h) * I moved docker.img & config paths to /mnt/cache (just wanted a bit of speed boost for Plex) * I created a tempfs 2GB disk within Plex docker container, mapped to /tmp (for Plex transcoding to RAM, I infrequently support 1-2 streams) * I enabled VLANs in Network settings (mostly a pre-emptive move as I'm moving in a month and slowly creating a new house network and wanted to see what VLANs feel like in Unraid, but this was supposed to be just for testing) Anyway, all changes went very smoothly, no errors, no unusual entries in logs. System felt snappier of course with 24GB RAM given I'd often be at 88-95% RAM used. But then I started to get a ton of notifications from my *arr containers that they can't see each other. Well, I figured I messed something up so I retraced my steps, changed everything back, but the problem persists. I then restored my appdata to a previous config just to be safe using CA Restore, updated containers, and rebooted, but I'm still seeing the same problem. Docker containers can't see each other. They're still on the same subnet, still same port assignments, still using Bridge mode. Oh, and I'm also using Binhex Deluge container with Privoxy, and even though the container starts up and I can see OpenVPN obtain an IP, privoxy doesn't see it and watchdog just blinks in an out, again, after running without a hiccup for ~2yrs. 2021-02-27 13:09:31,761 DEBG 'watchdog-script' stdout output: [info] Attempting to start Privoxy... 2021-02-27 13:09:31,760 DEBG 'watchdog-script' stdout output: [info] Privoxy not running 2021-02-27 13:09:31,761 DEBG 'watchdog-script' stdout output: [info] Attempting to start Privoxy... 2021-02-27 13:09:32,765 DEBG 'watchdog-script' stdout output: [info] Privoxy process started [info] Waiting for Privoxy process to start listening on port 8118... 2021-02-27 13:09:32,768 DEBG 'watchdog-script' stdout output: [info] Privoxy process listening on port 8118 2021-02-27 13:10:02,910 DEBG 'watchdog-script' stdout output: [info] Privoxy not running Any thoughts on how to troubleshoot what could've happened (and why undoing my steps didn't fix this?) Everything screams routing to me but I've no idea where to really go next.
February 27, 20215 yr Author For anyone with the same issue, looks like my little maintenance timing overlapped Binhex rolling out a potentially breaking change to fix a DNS leak. Issue solved after following his Q&A.
Archived
This topic is now archived and is closed to further replies.