[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: kworker eats 45% from first cpu



Jochen Spieker wrote:

> Jochen Spieker:
>> deloptes:
>>> 
>>> I moved my system to an SSD two days ago. I applied recommendation to
>>> use "discard"
>> 
>> What do you mean by that? If you are talking about the mount
>> option, discard it. (Sorry for the pun.) Use a weekly cron job with
>> 'fstrim -a' instead. If you are using LVM, set issue_discards=1 in
>> /etc/lvm/lvm.conf.
> 
> BTW, mount(8):
> 
>  discard
>         Disable/enable  the  discard mount option.  The discard function
>         issues
>         frequent commands to let the block device reclaim space  freed  by
>          the
>         filesystem.   This  is  useful for SSD devices, thinly provisioned
>         LUNs
>         and virtual machine images, but  may  have  a  significant 
>         performance
>         impact.   (The fstrim command is also available to initiate batch
>         trims from userspace.)
> 
> J.

Thank you Jochen. Yes I mean the discard option as you mentioned above, but
I do not think it is related. I configured it by the book
(https://wiki.debian.org/SSDOptimization)

I read in another forum it is not recommended to use discard in /etc/fstab.
I took the cronjob/weekly option.

I just mentioned it for completeness. Nohting else was changed on the system
and this issue was not there before the problem observed (before moving to
the ssd).

I think it might be something hanging on boot. Could be that something was
broken when moving the system - I used the classical tar cf - | (cd ...;
tar xf -) method. I can't imagine it is something wrong with it.

thank you in advance


Reply to: