Another node is killing my VPS

[eth72]

I've been with Jag for almost a year now and reasonably happy with the pricing, stability and support. However, about a few weeks ago, I started having unexplainable errors--memory warnings, CPU maxed out, system files suddenly not found, etc.
I contacted support and after a few days thinking it was my problem, realized that another node had a script or something that was robbing the entire CPU and severely affecting mine. They finally disabled the troubled node over the past weekend and my sites starting running optimally again.
3 days later (I'm assuming they started the trouble node back up), I start getting more alerts that MySQL is down and being restarted, SPAMD being restarted, etc. I receive an email from one of my customers complaining that the site was slow again last night. I check the VPS logs, and sure enough, on many occasions the CPU, etc., are maxed out.
My sites use very low bandwidth/usage and shouldn't hog resources. The load is almost always nothing.
Can anything be done to restore stability on triton.nocdirect.com, or do I have to live with it and start looking elsewhere? Personally, I don't want to lose my customers, nor leave JaguarPC...

[thisisit3]

Either ask support to re-examine the hardware node and stop the offending VPS (again), or ask them to move your VPS to another hardware node, thus you won't need to leave from Jag

[Frank Broughton]

as thisisit3 said, ask support to help - that is what they do. If you do not get satisfaction ask for JPC-Rizwan or JPC-NickO - I guarantee they will solve the problem. But you still must be diligent - they cannot control who puts what on a VPS.

[JPC-Rizwan]

Hi eth72,

We keep on blocking any resource abusive scripts or vpss to make sure the node or server remains stable , though in some scenarios it can take some time to catch the root cause.
Please let me know your ticket number and I will get it checked.

[thecoalman]

How would one determine what is an acceptable level of CPU usage for a VPS, unless I'm mistaken the CPU meter listed in either the Virtuozzo panel or using top in SSH is for the whole VPS so you really can't go by that or am I wrong?

The point to my question is I don't mind paying for what I need but I certainly don't want to wake up some morning to find my sites shut down because of an extended escalating problem.

Is there anyway to monitor it on a per VPS usage?

[eth72]

Thanks all for your reassurance... According to Virtuozzo, the CPU is back at 100%, which delays my SSH sessions terribly, and any php based websites sit there for a minute before starting to load. I don't mind the fact that the CPU(s) get shared which is the beauty of a VPS--but when it's hitting 100% from someone else and affecting my server, I'm potentially losing customers all around--not good... Thanks, too, Rizwan, I look forward to hearing from you...
Scott

[JPC-Rizwan]

Hi Scott,

I received your pm , thanks for providing the ticket number. I am checking into this and will update your ticket soon with findings.