| View previous topic :: View next topic |
| Author |
Message |
Bones McCracker
Veteran
Joined: 14 Mar 2006
Posts: 1611
Location: U.S.A.
|
 Posted: Thu Apr 05, 2007 9:45 am Post subject: right fstab "dump" / "pass" settings for |
 |
|
The namesys web site says reiserfs should have 0 and 0 for dump and pass.
http://www.namesys.com/faq.html#fstab
I have been running it with 1 for root partition and 2 for the others. No problems. I've seen it properly recover transactions following a freeze and forced reset. And there is a reiserfsck, although I know the filesystem is different in its use of metadata journalling.
So, is their website correct?
Does that apply to all the partitions?
This isn't a poll of what settings everyone is using. I'd like to hear from someone who actually understands how this filesystem is different.
Thanks. 
Last edited by Bones McCracker on Sun Apr 29, 2007 7:36 pm; edited 1 time in total |
|
| Back to top |
|
 |
timeBandit
Bodhisattva
Joined: 31 Dec 2004
Posts: 2719
Location: here, there or in transit
|
| Posted: Thu Apr 05, 2007 2:42 pm Post subject: |
|
|
| The namesys FAQ wrote: | What should I put into the fifth (aka dump, fs_freq) and the sixth (aka pass, fs_passno) fields of /etc/fstab for ReiserFS filesystems?
Why are ReiserFS filesystems not fscked on reboot after a crash?
Because ReiserFS provides journalling of meta-data. After a crash, the consistency of a filesystem is restored by replaying the transaction log.
Can I interactively repair a filesystem that was corrupted (due to an internal bug in the kernel or a to hardware fault)?
man reiserfsck |
They're not entirely correct.
Journaling does not completely eliminate the need for filesystem checks. Journaling dramatically reduces the time required to check and repair a dirty filesystem but the journal alone cannot repair all damage. In the vast majority of cases it may indeed be sufficient but not always.
The third question indirectly acknowledges this, and directly contradicts the second: a utility exists to repair damage that is not fixable using the journal alone. Naturally such damage could never occur during a crash, so it makes no sense to run an automatic check after a restart. 
The dump (fs_freq) column should be 0 because dump only works with ext2/ext3 filesystems and is generally a lousy backup tool anyway.
Your pass (fs_passno) settings reflect the "conventional wisdom" and are just fine.  The main drawback would be that you'll waste a little time during startup, performing filesystems checks that are rarely necessary.
_________________
Plants are pithy, brooks tend to babble--I'm content to lie between them.
Super-short f.g.o checklist: Search first, strip comments, mark solved, help others. |
|
| Back to top |
|
|
Bones McCracker
Veteran
Joined: 14 Mar 2006
Posts: 1611
Location: U.S.A.
|
| Posted: Fri Apr 06, 2007 12:56 am Post subject: |
|
|
Thank you.
Yours are the same observations I made when I read that page (I was surprised to see them recommending no fsck and thought those two statements were contradictory).
I wouldn't be surprised to see erroneous advice like that on a "community" page, but that appeared to be a FAQ page directly from the source. On the other hand, elsewhere in the same FAQ document, maybe I should not be surprised given there is other erroneous information there (DeskStar model mentioned is not the one that was problematic).
I appreciate the time you took to actually consult the page and see what I was referring to. |
|
| Back to top |
|
|
cyrillic
Watchman
Joined: 19 Feb 2003
Posts: 7313
Location: Groton, Massachusetts USA
|
| Posted: Fri Apr 06, 2007 4:37 pm Post subject: |
|
|
My observation is that up until recently, reiserfs would do its own fsck every time a filesystem was mounted. This resulted in long mount times for large filesystems (60+ seconds for a 1TB filesystem). I think this is why there was no need to do an additional fsck at boot time.
Ever since kernel 2.6.19, reiserfs mount times have been much quicker, so my guess is that the builtin fsck was removed, and doing it at boot time may be a good idea, now. |
|
| Back to top |
|
|
Bones McCracker
Veteran
Joined: 14 Mar 2006
Posts: 1611
Location: U.S.A.
|
| Posted: Mon Apr 09, 2007 4:39 pm Post subject: |
|
|
As I'm looking at the bootup messages stream by, I think it is doing a separate check first, and then what may very well be a redundant one at automount time.
[Edit]: after observing this for a week and researching a bit more, I've decided to keep the standard pass entries (1 for root and 2 for others). |
|
| Back to top |
|
|
|
Installing Gentoo
