[kernel disk scheduler] tasks that use disk heavily starve other tasks of disk access
Bug #176370 reported by
N7DR
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux (Ubuntu) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
There are several ways to demonstrate this problem. This is an easy one:
1. On a disk with lots of space, create a large file X of size ~8GB
2. Issue the command "split -b 4000m X"
This will keep the disk occupied nicely. Until the job finishes, even a dual-core machine will respond erratically. Some jobs will be fine, while others will freeze for long periods of time. I believe that the difference between freezing and non-freezing jobs is that the jobs that are freezing are being greatly starved of even small amounts of time needed to perform disk accesses.
I experience this issue on 64-bit gutsy on a dual-core system with a 500GB drive that is about 50% full.
Changed in linux: | |
status: | Invalid → New |
To post a comment you must log in.
Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in it recently. We were wondering is this still an issue for you? Can you try with latest Ubuntu release? Thanks in advance.