Pile of Disks Server - handling fsck
by linuxbird from LinuxQuestions.org on (#6J8MZ)
A "pile of disks" server is setup on an old system, with a large power supply, and for starters it has 60 Tb of WD Red drives.
This system will be used intermittently, and sometimes will not be turned on except perhaps once a month. Since fsck on larger file systems take copious time, and since there are several large file systems, it would be nice to have a strategy. It seems that running fsck at boot time may not be optimal for all the filesystems. I am considering having fsck run periodically (by count or time) on some of the more frequently used filesystems. But the less frequently used ones would rely on a manual script run to fsck the filesystem sporadically, or at times when those filesystems will not be used.
Ext4 is being used presently.
Is there any conventional wisdom on how to handle very large filesystems?
This system will be used intermittently, and sometimes will not be turned on except perhaps once a month. Since fsck on larger file systems take copious time, and since there are several large file systems, it would be nice to have a strategy. It seems that running fsck at boot time may not be optimal for all the filesystems. I am considering having fsck run periodically (by count or time) on some of the more frequently used filesystems. But the less frequently used ones would rely on a manual script run to fsck the filesystem sporadically, or at times when those filesystems will not be used.
Ext4 is being used presently.
Is there any conventional wisdom on how to handle very large filesystems?