No more ext4 maximal mount count

Those of us who have been at this for a while will have many fond memories of the "/dev/foo has reached maximal mount count" boot-time message, followed by a time-consuming full check of the filesystem in question. The recollection of times when one was standing in front of a room full of people and already late to start a presentation brings a special sort of joy. But it's likely that few of us remember the last time we saw such a message on a newer ext4 filesystem; now the documentation is catching up.

The mount-count check was there to force an occasional fsck run just in case some silent corruption might have found its way into the filesystem. The tune2fs command has had the ability to disable these checks since 1993, but the man page has long admonished against it:

The only problem here is that the mount-count-dependent checking was disabled by default in 2011. Or, as Eric Sandeen put it: "We did 'strongly consider the consequences' and disabled it by default". On the theory that there is "no need to scare the user about it now", he has proposed that this text be removed from the man page in favor of gentler text suggesting that some users may want to turn the feature back on. One suspects that most of us, though, are happier without random fsck delays; the more worried among us would probably rather schedule regular checks at predictable times.

---

to post comments