Memory Problems after Change on VPS servers

[byoungdt]

Today JPC switched the memory allocation for our VPS from UBC to SLM at about 3:10 pm PDT. Prior to the switch we had no major memory problems. If we exceeded our soft limits for short durations nothing happened. Now after the switch processes on our VPS are randomly killed, and worse the server is going down even if we are under the soft limit. If the soft limit is exceeded at all processes are killed none including init seem to be exempt. Our mta our webserver etc. all have been killed since 3pm.

I doubt that I am the only one having issues.... Has anyone else experienced random process kills and server crashes this afternoon and evening post "upgrade". Support is saying everything is fine.

[byoungdt]

Well I found the answer to my previous post a bunch of others are having the same issue and some are moving to new hosts:

http://www.webhostingtalk.com/showthread.php?p=4658042

[EuroNut]

Today JPC switched the memory allocation for our VPS from UBC to SLM at about 3:10 pm PDT. Prior to the switch we had no major memory problems. Support is saying everything is fine.

I had this issue a few days ago and Support fixed it within a short space of time. As I spotted something going wrong I added a message to the existing ticket so that anyone picking up the ticket could see the history of what had already been done. In fact, the ticket is still open but I have had zero issues since 10th August so I will probably close the ticket this weekend.

Well I found the answer to my previous post a bunch of others are having the same issue and some are moving to new hosts:

I can't help but wonder why some people are so keen to "jump ship" so quickly when something like this occurs. I have only been with JPC since April 2007, and had a couple of bumpy moments, but their reputation (which was what brought me to them in the first place) is good, and support's reaction to my (sometimes dumb) questions has always been prompt, efficient and effective.

My advice to anyone thinking of leaving is to give support a chance to get to grips with this, and keep the same ticket running till you get it fixed.

Dave (A happy JPC newbie)

[Ron]

Hi Dave!

[Vin DSL]

I can't help but wonder why some people are so keen to "jump ship" so quickly when something like this occurs...

My advice to anyone thinking of leaving is to give support a chance to get to grips with this, and keep the same ticket running till you get it fixed...

Good advice!

[byoungdt]

I can't help but wonder why some people are so keen to "jump ship" so quickly when something like this occurs.

Well if you run a blog then no big deal. We sell computer components online. Downtime means lost sales and even worse lowered trust. For sometime when we were on shared we actually were using a vps at another host to run our site since the shared we were on was oversold. We then ran parallel with a vps at JPC everything was fine. We figured that vps was going to be a solution until yesterday at 3:10pm PDT when they switched to SLM. Now the sever crashes and kills processes left and right. It has become unusable....

[EuroNut]

I can't help but wonder why some people are so keen to "jump ship" so quickly when something like this occurs.

Well if you run a blog then no big deal. We sell computer components online. Downtime means lost sales and even worse lowered trust.

Oh please don't think I don't appreciate the problem this can cause, nor the importance of good up time. For the record, the VPS I had issues with runs a hosting site, 3 on-line stores, an ad-server which feeds advertising off to 7 or 8 other servers in the US and Canada, 2 or 3 blogs, and some other stuff. It was a PITA for me to have any of that off-line.

We figured that vps was going to be a solution until yesterday at 3:10pm PDT when they switched to SLM. Now the sever crashes and kills processes left and right. It has become unusable....

Which is precisely why I made the suggestion to get a ticket running and start fixing things fast with JPC Support. The alternative of moving to another HSP (again, in your case, and several other by the sound of it) just seems to me to be playing hopscotch around hosts until you settle down on one where your VPS doesn't crap-out for some reason for a couple of months.

This whole swing from UBC to SLM is common throughout the industry, and some HSPs who overload their servers are fast going to come a major cropper when they try to follow suit.

I presume you did some research around the industry before settling on JPC? So I presume you picked them for the right reasons like they don't overload servers, they do use high-end machines, they do throw a lot of quality RAM into their boxes, and they don't rip you off price-wise.

Sorry if I'm ranting a bit - I just don't get this whole "let's swap hosts every month" philosophy one bit.

Good luck with whatever you decide

Dave

[JPC-Masood]

All those having SLM issues, we are moving you back to UBC parameters tonight.

[EuroNut]

All those having SLM issues, we are moving you back to UBC parameters tonight.

But, but, but! .....

What if we don't WANT to be moved back? (Just kidding!)

Dave

[Vin DSL]

But, but, but! .....

What if we don't WANT to be moved back?

Hahahaha!

Isn't that the way it always goes?!?!?!

[Medhat]

All those having SLM issues, we are moving you back to UBC parameters tonight.

I was on Appollo VPS plan just a month ago
and I switched to Enterprise plan as my resouces was going out about once or twice a month
but since I switched my RAM is out twice daily !!

Although before switching I had only 384 MB Guranteed Ram ,
after switching I suppose to have 768MB Guranteed Ram !

I host about 40 websites , half of them are not active , 0 visitors a month
and the other half is just normal websites .

I contacted the support they said that I have to buy more ram !
I don't know what to do !


if moving back to UBC will solve this then please Mr.Masood get my vps back to it .

[lamnk]

Too bad switching back to UBC doesn't solve the problem (at least for me, don't know about the others). My httpd is still killed all the time.

[JPC-Veena]

lamnk,

Since slm is not even enabled, vz cannot kill http on your vps. Please open a support ticket.

[EuroNut]

I was on Appollo VPS plan just a month ago
and I switched to Enterprise plan as my resouces was going out about once or twice a month
but since I switched my RAM is out twice daily !!

Although before switching I had only 384 MB Guranteed Ram ,
after switching I suppose to have 768MB Guranteed Ram !

if moving back to UBC will solve this then please Mr.Masood get my vps back to it .

Have you logged into your Virtuozzo control panel and looked at the QoS alerts?

If you can pin down a time when things go wrong, and see what your QoS screen is saying at that time, you'll be half-way there. Then just open a support ticket with the relevant info ....

[Medhat]

my problem is always with privvmpages
I rarely face any problem with any other thing .