Ways to avoid OOMKiller problem?
Hi, We came across this - https://www.monetdb.org/Documentation/OOMkiller and we were thinking on some ways to overcome this situation. 1. Is there any configuration in MonetDB to restrict the maximum RAM usage of the mserver process? 2. Is there any configuration in MonetDB to restrict the maximum RAM used by a single thread? 3. Is there any suggested solution for the OOMKiller problem? If you have explored anything else on this, please share your experience with MonetDB server Memory management. Any help much appreciated. Thanks in advance. Regards, Vijay.
Hi On 28/03/16 13:23, Vijay Krishna wrote:
Hi,
We came across this - https://www.monetdb.org/Documentation/OOMkiller and we were thinking on some ways to overcome this situation.
1. Is there any configuration in MonetDB to restrict the maximum RAM usage of the mserver process? no there is none 2. Is there any configuration in MonetDB to restrict the maximum RAM used by a single thread? no there is none 3. Is there any suggested solution for the OOMKiller problem? Actions have been taken within the kernel to avoid it as best as possible. regards, Martin
If you have explored anything else on this, please share your experience with MonetDB server Memory management.
Any help much appreciated. Thanks in advance.
Regards, Vijay.
_______________________________________________ users-list mailing list users-list@monetdb.org https://www.monetdb.org/mailman/listinfo/users-list
Here: http://unix.stackexchange.com/questions/153585/how-oom-killer-decides-which-... is a description of how the OOMKiller decides which process to kill. It also describes the possibility of adjusting the "badness score" of a process, even as far as making it impossible for OOMKiller to kill (although I'm not speaking from experience, just quoting). Vijay, have you tried that? Martin/other developers - have we played around with it? Eyal On 28/03/2016 21:02, Martin Kersten wrote:
Hi
On 28/03/16 13:23, Vijay Krishna wrote:
Hi,
We came across this - https://www.monetdb.org/Documentation/OOMkiller and we were thinking on some ways to overcome this situation.
1. Is there any configuration in MonetDB to restrict the maximum RAM usage of the mserver process? no there is none 2. Is there any configuration in MonetDB to restrict the maximum RAM used by a single thread? no there is none 3. Is there any suggested solution for the OOMKiller problem? Actions have been taken within the kernel to avoid it as best as possible. regards, Martin
If you have explored anything else on this, please share your experience with MonetDB server Memory management.
Any help much appreciated. Thanks in advance.
Regards, Vijay.
_______________________________________________ users-list mailing list users-list@monetdb.org https://www.monetdb.org/mailman/listinfo/users-list
_______________________________________________ users-list mailing list users-list@monetdb.org https://www.monetdb.org/mailman/listinfo/users-list
Thanks Martin/Eyal for the replies.
I haven't tried with the "badness score" adjusting and I prefer to have
this as the last option.
Rather, I would like to prevent my system from being running out of memory
by restricting the RAM and Disk usage of MonetDB. That is why I was asking
if there is any builtin configurations with MonetDB for the same.
I read that I could also restrict the usage with ulimit.
But is that advisable? Will such restrictions create any side effects like
crash or restart, other than impacting the query performance?
Thanks & Regards,
Vijayakrishna.P.
Mobile : (+91) 9500402305.
On Tue, Mar 29, 2016 at 12:25 AM, Eyal Rozenberg
Here:
http://unix.stackexchange.com/questions/153585/how-oom-killer-decides-which-...
is a description of how the OOMKiller decides which process to kill. It also describes the possibility of adjusting the "badness score" of a process, even as far as making it impossible for OOMKiller to kill (although I'm not speaking from experience, just quoting).
Vijay, have you tried that? Martin/other developers - have we played around with it?
Eyal
On 28/03/2016 21:02, Martin Kersten wrote:
Hi
On 28/03/16 13:23, Vijay Krishna wrote:
Hi,
We came across this - https://www.monetdb.org/Documentation/OOMkiller and we were thinking on some ways to overcome this situation.
1. Is there any configuration in MonetDB to restrict the maximum RAM usage of the mserver process?
no there is none
2. Is there any configuration in MonetDB to restrict the maximum RAM used by a single thread?
no there is none
3. Is there any suggested solution for the OOMKiller problem?
Actions have been taken within the kernel to avoid it as best as possible. regards, Martin
If you have explored anything else on this, please share your experience with MonetDB server Memory management.
Any help much appreciated. Thanks in advance.
Regards, Vijay.
_______________________________________________ users-list mailing list users-list@monetdb.org https://www.monetdb.org/mailman/listinfo/users-list
_______________________________________________ users-list mailing list users-list@monetdb.org https://www.monetdb.org/mailman/listinfo/users-list
_______________________________________________ users-list mailing list users-list@monetdb.org https://www.monetdb.org/mailman/listinfo/users-list
participants (3)
-
Eyal Rozenberg -
Martin Kersten -
Vijay Krishna