Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230

 

 

 

Disk IO Degradation over long periods of high IO

Here you can discuss every aspect of Debian. Note: not for support requests!
Post Reply
Message
Author
cjq
Posts: 2
Joined: 2017-02-03 00:46

Disk IO Degradation over long periods of high IO

#1 Post by cjq »

Hey, I have had this issue for years when copying large amounts of data between disks.

Initially you get high io speeds and the disk cache will max out from there and then the system will become sluggish and eventually unresponsive and the copying slows down to almost zero, this happens on any debian system(although it may happen on other distros but I generally exclusively use debian when doing this task).

I do have a work around which I run in a bash root shell when doing large copies which maintains the the hi throughput copying which constantly does a sync and drops the caches:

Code: Select all

while true ; sync ; echo 3 > /proc/sys/vm/drop_caches ; done
This has to be started before the copying as once it becomes sluggish it never seems to recover until after a reboot.

Just wondering if anyone has encountered this and has found a better solution to resolving this? I have tried tuning the kernel parameters around caching but it doesn't seem to make a difference.

Thanks

User avatar
Head_on_a_Stick
Posts: 14114
Joined: 2014-06-01 17:46
Location: London, England
Has thanked: 81 times
Been thanked: 132 times

Re: Disk IO Degradation over long periods of high IO

#2 Post by Head_on_a_Stick »

Try a newer kernel, perhaps?

https://lwn.net/Articles/682582/
deadbang

cjq
Posts: 2
Joined: 2017-02-03 00:46

Re: Disk IO Degradation over long periods of high IO

#3 Post by cjq »

I'm already running the backport 4.8 Kernel on these where I am experiencing the issue although as I have said have experienced this through the 3.* series kernels too. Just trying out the 4.9 from today.

Thanks for the link, that appears to be the same issue... however no resolution.

Post Reply