January 26, 201511 yr unRAID OS Version: 6.0 beta 12 Description: ReiserFS brings down system. How to reproduce: Start MC in a screen session. Move more files than will fit onto a mature ReiserFS disk. By mature, I mean a large ReiserFS formated disk that has been in use for a while, been close to full capacity several times. Expected results: System should still be reachable by telnet. Actual results: System is never heard from again. Webgui/unmenu unreachable. No response to telnet. However, system can still be pinged. Other information: ReiserFS clearly performs worse on 6b12 than it did on 5.05. The issue described here has fairly unpleasant effects. Guess why there's no syslog attached.
January 26, 201511 yr Thats a known issue with Reiser. As it gets very close to being filled, the delays it imposes trying to find free space becomes longer and longer. All file systems have the same issue to a lesser or greater extent, but Reiser does seem to be bad for it. Not sure how XFS is
January 26, 201511 yr One of my servers has 21 almost full drives, with 5.0.5 it was a pain to add or even move files, kept getting time outs although the server didn't stop responding. My primary reason to upgrade to 6b12 was to try XFS because other forum users said it was much better with full disks, took me 15 days to convert all discs but it was definitely worth it, adding more files or moving is now almost instantaneous.
February 9, 201511 yr Author I am currently transitioning to XFS. As part of the process I wanted to generate blake2 hashes with the bunker tool, but with several of my disks, I have encountered the reiserfs issue with cpu stalls every 3 minutes in syslog and unresponsive disks and unable to stop the array. I have had to hard power off and do a parity check every 3 or 4 days these past couple of weeks. Have only managed to convert 3 disks so far.
Archived
This topic is now archived and is closed to further replies.