LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)
-   -   Question about FSCK (https://www.linuxquestions.org/questions/linux-newbie-8/question-about-fsck-4175580241/)

pratik@7z 05-20-2016 03:40 AM

Question about FSCK
 
Can you provide last option which is FSCK order in /etc/fstab for NFS partition on NFS client or NFS server

michaelk 05-20-2016 06:53 AM

For the client the the fsck field is not used and should be zero.

For the server depends on the actual file system. If other then the /root 2 is typically used.

dab1414 05-20-2016 07:05 AM

Read this.

Basically, your clients will always have the last 2 fields (dump,fsck) at 0. And your server depends on your needs. Do you want FSCK to run at every boot? So root should be 1 then other partitions you want to fsck a 2.

jpollard 05-21-2016 09:07 AM

Quote:

Originally Posted by dab1414 (Post 5548162)
Read this.

Basically, your clients will always have the last 2 fields (dump,fsck) at 0. And your server depends on your needs. Do you want FSCK to run at every boot? So root should be 1 then other partitions you want to fsck a 2.

well... actually that last parameter specifies the ORDER fsck is to be run.

0 is a default.
1 normally used at boot for the root filesystem only.
2 all filesystems with a "2" will be done in parallel, and AFTER any filesystems with a lower number are completed.
3 all filesystems with a "3" will be done in parallel, and AFTER any filesystems with a lower number are completed...

and so on.

With the common uses of journaled filesystems, this value has become "mostly" redundant, and may be 0 even on a server. If a filesystem is "unclean" (crash/power failuer...) all that is done is that the journal is processed to finish the last active transactions, and then mounted.

The reason this is being done is that with multi-TB filesystems, it can take HOURS (I have seen it take DAYS) to perform an fsck.

Journal based filesystems eliminate this by eliminating the corruption caused by crashes. The transactions are either completed, or rolled back, and thus keeping the filesystem consistent without needing a detailed scan.


All times are GMT -5. The time now is 06:06 PM.