Re: prevent fs check after being up for a long time




I had to reboot a server today that has been up for about 300 days


Why did it need a reboot?


It shutdown was clean and so i would expect the reboot to be pretty quick.
Issue was that i was told that as a particular filesystem has not been
fs checked for 200 and something days it was going to check it. This
filesystem is a couple of terrabytes and took ages!


The normal philosophy is to assume that there will be something go wrong
after such a long time. If you are not maintaining the system - keeping
the critical stuff up to date, this is really a good idea.


Is there anyway to stop this from happenning?


Several ways. If you feel that the check isn't going to be needed, you
can use a tool to turn off the check. An example would be tune2fs - see
the man page. Another solution when you are going to reboot the system
would be to touch a file named '/fastboot' which rc.sysinit looks for to
bypass the file checks.

A better solution would be to periodically (during slack periods of use)
remount the file system 'read-only' and do a quicky fsck. If nothing is
discovered, remount 'read-write' and you're home free. If it does find a
problem, have it mail you so that you can schedule a repair. This solution
does require some thinking, and a lot depends on how your file system is
arranged. DO NOT CHECK A FILE SYSTEM THAT IS MOUNTED READ-WRITE. Ideally,
the filesystem should be unmounted, though this is often a problem.

The best solution is to schedule maintenance. You don't mention a release,
but 300 days says at least FC3, and it's had a least two kernel errata in
those 300 days. Even the virtually unsupported RH9 has had two kernel errata
in the same period.

thanks for the tips - i will investigate what you suggested. However to answer your main question regarding why this needed a reboot i had to remove a SCSI tape drive from the back of the box and so had to halt it to do it. This is a LAN based fileserver and is not internet connected and so i'm not that urgent about applying kernel updates to it. Its a fileserver basically and nothing else so there is no user access so local root exploits also so not really apply.

thanks for the tuning tips i'll take a look at them
.

Relevant Pages

  • Re: prevent fs check after being up for a long time
    ... Issue was that i was told that as a particular filesystem has not been ... Another solution when you are going to reboot the system ... remount the file system 'read-only' and do a quicky fsck. ... Even the virtually unsupported RH9 has had two kernel errata ...
    (linux.redhat)
  • Re: filesystem full error with inumber
    ... The file system is running out of data space. ... the reboot. ... Those have nothing to do with "filesystem full". ... Such as df not reporting the actual data usage, ...
    (freebsd-stable)
  • Re: fsck and memory filesytems (fsck_mfs: No such file or directory)
    ... but I still get the same behaviour next reboot. ... the order in which file system checks are done at reboot time. ... before checking some filesystem, you ensured that its mount point was ... Let me point to the same lines of manfstab. ...
    (freebsd-questions)
  • FILESYS: Waiting for bootable file system to be mounted.
    ... File System" as my filesystem and ROM-only File System, ... however, when reboot, I got "FILESYS: ...
    (microsoft.public.windowsce.platbuilder)
  • SUMMARY: how do I REALLY delete a file?
    ... leaving the rest of the file system intact. ... Wipedrive doesn't seem to be available for Solaris, but might be of interest to ... run this on each filesystem where the files from ... >Solaris' UFS] do not satisfy this assumption." ...
    (SunManagers)