[Check_mk (english)] tmpfs will not release (no free space)

k19840 at 126.com k19840 at 126.com
Wed Dec 24 12:43:43 CET 2014


hi Andreas,

thanks. list as below

[root at monitor tmp]# ls 
apache check_mk dokuwiki icinga lock nagios nagvis php pnp4nagios rrdcached rrdcached.pid run shinken thruk 
[root at monitor tmp]# du -h 
0 ./nagvis/share 
0 ./nagvis/tmpl/cache 
0 ./nagvis/tmpl/compile 
0 ./nagvis/tmpl 
0 ./nagvis 
0 ./apache/fcgid_sock 
4.0K ./apache/run 
8.0K ./apache 
0 ./thruk/reports 
0 ./thruk 
0 ./rrdcached 
0 ./php/wsdl-cache 
27M ./php/session 
0 ./php/upload 
27M ./php 
24K ./nagios/checkresults 
0 ./nagios/tmp 
584K ./nagios 
0 ./shinken/run 
0 ./shinken/tmp 
0 ./shinken 
768K ./check_mk/cache 
152K ./check_mk/counters 
920K ./check_mk 
4.0K ./pnp4nagios/run 
0 ./pnp4nagios/lock 
4.0K ./pnp4nagios/stats 
8.0K ./pnp4nagios 
0 ./run/mysql 
0 ./run 
4.0K ./lock 
0 ./dokuwiki 
28M . 

[root at monitor session]# df -h 
Filesystem Size Used Avail Use% Mounted on 
/dev/mapper/vg_monitor-lv_root 
211G 5.9G 194G 3% / 
tmpfs 1.2G 0 1.2G 0% /dev/shm 
/dev/sda1 485M 55M 405M 12% /boot 
/dev/mapper/vg_monitor-lv_home 
9.9G 151M 9.2G 2% /home 
tmpfs 1.2G 29M 1.2G 3% /opt/omd/sites/hq/tmp 
tmpfs 1.2G 1.1M 1.2G 1% /opt/omd/sites/service/tmp 

in /php/session are full of files like these:
sess_58go55cg5u9tev1gnbcret8s81 sess_aima3kkt4iadlsv4af0p385en6 sess_g6hduao808p0kq039k87e8trf3 sess_l8ubmjsecrf4qgtgi54dchgvs5 sess_qg65de17ua09rmmk9ojb9pkv02 
sess_58n4m97e5jo2j5thu94fojvnu1 sess_ain65to2b9353o5jd71li1ljf1 sess_g6q8rthsfomm83g34kjtv8p0m5 sess_l8vvgaf2etk2apfcespa3fl0k7 sess_qg7fc4k8i05lr4nq4sevve6ia1 

PHP Version
[root at monitor session]# php -v 
PHP 5.3.3 (cli) (built: Aug 6 2014 05:54:27) 
Copyright (c) 1997-2010 The PHP Group 
Zend Engine v2.3.0, Copyright (c) 1998-2010 Zend Technologies 

BR,
KT



 
From: Andreas Döhler
Date: 2014-12-24 19:29
To: k19840 at 126.com; checkmk-en
Subject: Re: Re: [Check_mk (english)] tmpfs will not release (no free space)
That's no answer to the question what's inside. Please take a look inside the tmpfs and locate the files who consume your space. Then it is way easier to give some advice.
br
Andreas

k19840 at 126.com <k19840 at 126.com> schrieb am Mi., 24. Dez. 2014 12:19:
Hi Andreas,

just standard installation, and the VM(CENT OS 6.6 mini installation) only have OMD installed. no other script/app will touch that folder or store anything else over there.  

but really strange is hq site have less host got that problem, service site have more host but have no problem.

* just add more ram. ( to 2.5G). and saw the tmpfs size still counting in every minute.

[root at monitor ~]# top 
top - 19:07:31 up 1 min, 1 user, load average: 0.10, 0.06, 0.02 
Tasks: 203 total, 1 running, 202 sleeping, 0 stopped, 0 zombie 
Cpu(s): 2.3%us, 1.7%sy, 0.0%ni, 95.9%id, 0.0%wa, 0.0%hi, 0.1%si, 0.0%st 
Mem: 2413464k total, 469188k used, 1944276k free, 16040k buffers 
Swap: 4063224k total, 0k used, 4063224k free, 185840k cached 

[root at monitor ~]# df -h 
Filesystem Size Used Avail Use% Mounted on 
/dev/mapper/vg_monitor-lv_root 
211G 5.9G 194G 3% / 
tmpfs 1.2G 0 1.2G 0% /dev/shm 
/dev/sda1 485M 55M 405M 12% /boot 
/dev/mapper/vg_monitor-lv_home 
9.9G 151M 9.2G 2% /home 
tmpfs 1.2G 7.1M 1.2G 1% /opt/omd/sites/hq/tmp 
tmpfs 1.2G 1.1M 1.2G 1% /opt/omd/sites/service/tmp 

* 1 minute later

[root at monitor ~]# df -h 
Filesystem Size Used Avail Use% Mounted on 
/dev/mapper/vg_monitor-lv_root 
211G 5.9G 194G 3% / 
tmpfs 1.2G 0 1.2G 0% /dev/shm 
/dev/sda1 485M 55M 405M 12% /boot 
/dev/mapper/vg_monitor-lv_home 
9.9G 151M 9.2G 2% /home 
tmpfs 1.2G 8.7M 1.2G 1% /opt/omd/sites/hq/tmp 
tmpfs 1.2G 1.1M 1.2G 1% /opt/omd/sites/service/tmp 
[root at monitor ~]# 

thanks everyone & regards
have a nice chrismas.
KT



 
From: Andreas Döhler
Date: 2014-12-24 18:15
To: k19840 at 126.com; checkmk-en
Subject: Re: [Check_mk (english)] tmpfs will not release (no free space)
Hi KT,

beside what Marcel already mentioned, what is inside your tmpfs?
243MByte usage looks very high for such a small site.
Is it possible that you have other processes / scripts who store data there?

Best regards,
Andreas

k19840 at 126.com <k19840 at 126.com> schrieb am Wed Dec 24 2014 at 10:09:12:
Hi everyone,

just got a strange problem: tmpfs will not release, then stuck whole OMD (web page monitoring services turns grey, can't add new host). if i force omd service restart, apache will fail.
only OS reboot can fix it. ( for 1 day), the tmpfs folder will eat up 100% in less than 1 day)

searched seems no one have same problem, is there anything i can do to fix this? or i need more ram?

thanks a lot.

my config:
OMD 1.2 x64
CentOS 6.6 X64, up to date
RAM: 2GB
hq site: 41 hosts (which will ran into problem, it's tmpfs will keep 100% after OS reboot for less than 1 day), monitoring windows(snmp)/network devices(snmp)/linux(agent)
service site: 61 hosts

[root at monitor ~]# df -h 
Filesystem Size Used Avail Use% Mounted on 
/dev/mapper/vg_monitor-lv_root 
211G 5.8G 194G 3% / 
tmpfs 939M 0 939M 0% /dev/shm 
/dev/sda1 485M 55M 405M 12% /boot 
/dev/mapper/vg_monitor-lv_home 
9.9G 151M 9.2G 2% /home 
tmpfs 939M 243M 696M 26% /opt/omd/sites/hq/tmp 
tmpfs 939M 3.0M 936M 1% /opt/omd/sites/service/tmp 

Regards,
KT



_______________________________________________
checkmk-en mailing list
checkmk-en at lists.mathias-kettner.de
http://lists.mathias-kettner.de/mailman/listinfo/checkmk-en
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mathias-kettner.de/pipermail/checkmk-en/attachments/20141224/f2d02a2f/attachment.html>


More information about the checkmk-en mailing list