I am using a digitalocean lamp stack on ubuntu 14.04.2 and the free version of webmin - virtualmin
Some details about the issue i am facing:
for 3-4 weeks now login page crashes
in order to get acess i eventually stop and restart webmin from command line and some times this fixes the issue and i get access.
if above solution doesnt solve my issue i perform a reboot -h and then i finally get acesss.
I am very concerning about what is going on due i have to do this 2-3 times a week now.
Also today vps crashed - sites went down and couldnt login neither via putty or via ftp fillezilla.
CPU usage peeked at 100%. at that time.
Some Quotas :
40GB storage capacity using arround 9GB
2GB Memory using around 1,1GB
Problem solved only via digital ocean panel and power cycle option.
How can i solve this issue.
I didnt proceed to any update though i see some update notices for some time now.
Also in var/webmin/miniserv.error i see these errors :
Error: Failed to query Postfix config command to get the current value of parameter DOMAIN:
Error: Failed to query Postfix config command to get the current value of parameter recipient_bcc_maps:
Hello Eric,
I see that in my Digital Ocean VPS i could install swap file but they say better not do that, in SSD disks as mine.
Mostly they suggest an upgrade.
I have on that VPS 14 sites now (mostly wordpress sites) also i ve istalled all webmin-virtualmin updates but still i am facing problems - evey 2-3 days vps crashes and i proceed with a power cycle.
I was wondering if i could do anything else - maybe upgrade to virtualmin pro or i d better move some sites from that vps to a new one.
Any suggestions?
Thanks
George
VPS with 2 GB of ram should be enough for many things but all comes down to did you fine tune MySQL and Apache and after that what kind of sites are hosted, static pages, CMS and so on…
Just to give you an example, i have one VPS for testing with 1.8GB ram with 1.2GB swap and i did a test how much i can pull on WP home page. I done a test for 1 hour hammering the server with 3 clients/visitors per second (total of 10800 visitors/hour) and that was the limit what my VPS could hold without errors, once i raised to 4 clients per second things start to break with around 1-2% of connections timed out. Probably i could spend some time and push to higher level but there is no need for that, if you website get so many hits you must go for dedicated with SSD or at least KVM with 6-8GB.
For MySQL i would suggest to download mysqltuner to see what results you get and for Apache check the log files maybe some problems are asking for your attention. Default installation does nothing to help you out and few things are “broken” even on fresh install. For example, on 2GB VPS if you select MySQL with 1GB (post install wizard) it will still use more than 2GB of your total memory and if you leave it as it is you can soon expect problems. Doesnt help knowing there are no “one settings for everything” and even between two equal VPS your MySQL settings will be different. So it all comes down to test and modify on the fly.
TL;DR: Download mysqltuner to fine tune your DB, check Apache logs and if needed make adjustments.