Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 955
  • Last Modified:

Questions about jfs2 log.

OK, here he go wmp..

1- Which is better to use, INLINE or JFS2LOG?
2- Is possible to see the usage of the JFS2 log lv?
3- What happen if I umount a /filesystem, logform /dev/hd8 and mount /filesystem? The data is lost?

Not clear at all about how JFS2 log works..;)
0
sminfo
Asked:
sminfo
  • 3
  • 2
1 Solution
 
woolmilkporcCommented:
Hi, and welcome back in our new year 2013 - hope you had a good start!

1) Hard to say. Some people prefer having a distinct log per filesystem, and with this setup an inline log is much easier to handle. With inline logging there is a distinct log per FS by design, no need to create lots of new logvolumes.
On the other hand, if you have an FS with heavy metadata activity (create/delete/update) you could consider using a separate logvolume on a separate disk for performance reasons.
Anyway, a logvolume should be mirrored if feasible, and an outline log can have multiple copies even if the original FS is not mirrored (well, a rare scenario, I assume ...).
If an FS with an inline log is mirrored the "log" part is automatically mirrored too.

I, for one, tend a bit more to inline logs ...

2) If with "usage" you mean "filling level" - no you can't see that. Only the FS itself is aware of what is placed where in the log volume.
If you mean "activity" instead - this can be monitored with "lvmstat" just as any other logical volume.

3) Formatting a log with logform destroys the log records contained therein, regardless of the log type used. Formatting an outline log volume means reformatting the whole volume, running logform against a filesystem volume containing an inline log means reformatting only the "log" part of that volume which doesn't affect filesystem data.

A filesystem journal keeps track of changes to metadata, such as the superblock, inodes, pointers, directories. In case of a filesystem crash the information contained in the journal is used to recover this metadata.


wmp
0
 
sminfoAuthor Commented:
here I am wmp.. as always, with my topas at 100% USED...;)

OK, I understood everything.. thanks...
0
 
sminfoAuthor Commented:
BTW, do you have EMC storage there? I have a problem (case on IBM now) about mksysb of LPATRs and the EMC's powerhdisks and multipath.... tell me and I open a new question...;)
0
 
woolmilkporcCommented:
Sorry, no EMC anymore. And even in those old days we never had a rootvg on it.
0
 
sminfoAuthor Commented:
heheh... I'm suffering with this )*(&*)()&PJM EMC.....:-(

see ya!
0

Featured Post

What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now