2

Before a grow on a 8 TB RAID5, I am currently doing an e2fsck.

While the RAID was running w/o issues before (I switched from Ubuntu to CentOS 5.4 because I needed a specific hardware driver), the e2fsck initially turned up a boatload of errors (nohup.out is 7 GB), but after 5 days of continuous fscking, there have been no more outputs into the logs for nearly 2 days even though the process is still running at 100% CPU and 1.5G RAM. I tried an killall -USR1 e2fsck, but I still get no further outputs in my log.

Is it safe to kill e2fsck and retry with the -C option or will I make things worse? I have looked into the running times of e2fsck but from what I read by now it should have finished.

1 Answer 1

1

Until recently, I would have said that interrupting it was suicidal. However, five weeks ago, a colleague gaily interrupted a reboot fsck on a nastily dirty file system, before I could stop him. The FS was later fscked without incident. I have since, myself, interrupted several fscks on multi-TB file systems, because I didn't want them to happen yet, and I've not so far had any problems and when I redid them I used -C 0 as well.

As ever, Your Mileage May Vary, and whatever action you take is your own responsibility; but I was very surprised when my colleague got away with it, so I started to try it myself, and I've been alright so far. I thought that might be of interest to you.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .