Saturday, August 22, 2009

Fsck Big Drives

Most of downtown Toronto had a blackout this morning due to a fire in a transformer vault. My customer's servers in the area are not on a generator, so they shutdown once the UPS ran out. Everything came back up once power was restored but some of the Linux servers hadn't been rebooted in three months, which naturally forced an a fsck. Those servers have 500GB drives which seem to take forever to complete a file system check.

Unfortunately there's not much that can be done about that. Even with a journaled file system, it is recommended to fsck the drive regularly to detect failing drive media. But with big drives, the delay in boot time is kind of unnerving when you are not in front of the console where you can monitor the progress. And this is only going to get worst as we move to ever larger drive capacities.

No comments:

Post a Comment