Find the answer to your Linux question:
Results 1 to 5 of 5
Like Tree1Likes
  • 1 Post By docbop
Usually my filesystem is read only. So if I want modify it (delete file, add file) I remount it as rw. After modification I remount it as ro. Obviously this ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. #1
    Just Joined!
    Join Date
    Feb 2014
    Posts
    3

    jffs2 - save umount and mount


    Usually my filesystem is read only. So if I want modify it (delete file, add file) I remount it as rw. After modification I remount it as ro.
    Obviously this is not a good idea, as some internal processes are stopped.
    Is there a way to clean remount it (delayed), so that I can be sure, that everything is ok? As this runs on scripts, I can't just wait few minutes just on suspect.

    (Background: sometimes after modification I delete and reflash the hole partition and after mounting it I get some errors like this:

    Node CRC ffffffff != calculated CRC f09e7845 for node at 03b618c8
    read_cache_page() returned error: -5
    Error garbage collecting node at 03b618c8!

    This is strange to me, as I setup the whole partition. So the kernel seems to store some history about jobs to do and try to continue - even after wipe the partition.

  2. #2
    Linux Engineer docbop's Avatar
    Join Date
    Nov 2009
    Location
    Woodshed, CA
    Posts
    947
    What if you go old school. Back when I first got into Unix it was common to run the sync command a couple times before shutting down or other drive operations to insure all the buffers were flushed. I would try adding a couple sync command before your unmount.

    sync(: synchronize data on disk with memory - Linux man page
    voidpointer69 likes this.

  3. #3
    Just Joined!
    Join Date
    Feb 2014
    Posts
    3
    I know the sync command. I forgot to mention, that this has no effect here. Problem seems to me 2 processes, which becomes active (but with low prio) :
    sync_supers
    jffs2_gcd_mtd7
    despite sync. I assume these are background processes, which make wearlevel/garbage collection ... such kind of stuff.
    What I miss is command to do this with higher prio and a clear signal that this this is fixed.

  4. $spacer_open
    $spacer_close
  5. #4
    Linux Engineer docbop's Avatar
    Join Date
    Nov 2009
    Location
    Woodshed, CA
    Posts
    947
    I was looking for a command that would show what processes are holding up an umount. Check the fuser command in this article.

    UNIX / Linux umount Command Examples

  6. #5
    Just Joined!
    Join Date
    Feb 2014
    Posts
    3
    Thanks, I will check the processes accessing the filesystem.
    Currently the umount doesn't give a hint that something is wrong. I guess it is intended that jffs2 act like this. The issue becomes a problem, If I just erase and reflash this partition an mount it again.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •