tucansam Posted October 2, 2023 Share Posted October 2, 2023 This is now happening all the time. It used to be a regular (once or twice a month) occurance, but today it has now happened three times, and I have some idea why. Shares are available to Windows machines. After a particular amount of file I/O onto the array, all shares disappear, and /mnt/user becomes unavailable with errors at the shell prompt. /mnt/user: Transport endpoint is not connected Today I am doing a lot of maintenance on my media library, and sabnzbd is downloading several terabytes of data. It gets to a certain point, successfully, and then the aforementioned issue pops up. Stopping and starting the array has no effect; only a reboot will bring /mnt/user back online. Then sab will continue about its business until the next the time issue happens. Like I said, sitting here for the last couple of hours, its now happened three time. This has been an ongoing issue for me for at least a year, but today it is bad enough (and consistent) enough that I've had enough. Diags attached. ffs2-diagnostics-20231002-0945.zip Quote Link to comment
tucansam Posted October 2, 2023 Author Share Posted October 2, 2023 Have seen similar threads going back to 2020 indicating it may be the mover. Some users indicated that turning the mover off had no effect. Some believe it is an issue with NFS -- I cannot turn NFS off. Quote Link to comment
JorgeB Posted October 2, 2023 Share Posted October 2, 2023 It's this old issue: Quote Link to comment
tucansam Posted October 2, 2023 Author Share Posted October 2, 2023 Is there a known fix? Or just theories? Quote Link to comment
JorgeB Posted October 2, 2023 Share Posted October 2, 2023 Some workarounds discussed in that thread, but AFAIK no definite fix since there's also no way to consistently reproduce it. Quote Link to comment
tucansam Posted October 2, 2023 Author Share Posted October 2, 2023 I can consistently reproduce it. Its just not throwing anything in syslog AFAIK. Anywhere else I can be looking? Quote Link to comment
JorgeB Posted October 2, 2023 Share Posted October 2, 2023 Can you give the steps needed so that I or someone else can reproduce it? Quote Link to comment
tucansam Posted October 2, 2023 Author Share Posted October 2, 2023 I have 250GB of queue going in sab. I have Windows Explorer open as I peruse the filesystem cleaning up old files, fixing renaming errors sonarr or sab screw up, etc. Within 60 minutes of conducting both of these activities at the same time, the issue surfaces. Its now happened six or seven times today. It seems to be the consistent manipulation of the filesystem in some way (usually unpacking downloaded rar files is the instant at which it occurs). It occurs when Explorer is closed and sab is the only thing messing with the filesystem (although in fairness Plex is doing its constantly watching thing in the background, indexing newly downloaded files) Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.