OpenVZ provider pegged at 120KB/s throughput

[drmike]

Using a provider in the States who I have issues with.  It's an annual plan, so until the year does us part, I am trying to get use out of what was paid for,

Getting always speed cap of 120KB/s throughput.  Day, night, weekends, etc.   Sometimes it goes lower, but returns to this horrible speed.

I've contacted the company in the past and they've resolved the issue after 10 tickets and weeks of wait.  Claim a reboot of the container solved the issue.

I've just rebooted such and I get this:

wget -O /dev/null http://cachefly.cachefly.net/100mb.test

--2013-05-14 03:48:36--  http://cachefly.cachefly.net/100mb.test

Resolving cachefly.cachefly.net (cachefly.cachefly.net)... 205.234.175.175

Connecting to cachefly.cachefly.net (cachefly.cachefly.net)|205.234.175.175|:80... connected.

HTTP request sent, awaiting response... 200 OK

Length: 104857600 (100M) [application/octet-stream]

Saving to: '/dev/null'

 

 0% [                                                                                                                                                                                                    ] 273,668      119KB/s             

13% [=========================>                                                                                                                                                                          ] 13,937,808   114KB/s  eta 10m 58s

 

Any providers have an explanation for what might be going wrong.  It's a generic Debian installation.  The container is entirely unused/idle.

 

[Nick]

What network/location is it and when did you last have issues with the container?

 

[drmike]

Location is Chicago.

Issue with the container is non stop.  Been this way for several months, minus the one time the provider actually did something and made it work right for < 1 week.

 

[shovenose]

Can you run a traceroute to something else in Chicago... like ChicagoVPS.net or something?

 

[drmike]

Traceroutes are fine/normal.  Routes are what one would expect and hop latency is as expected.  Network, yeah, the provider is on Colocrapping's network, but the issue isn't the network.

 

[MannDude]

Traceroutes are fine/normal.  Routes are what one would expect and hop latency is as expected.  Network, yeah, the provider is on Colocrapping's network, but the issue isn't the network.

Well, that narrows it down to, well, about 100 providers... Haha.

 

[drmike]

I don't want to finger-gun-bang any provider  Trying to avoid that sort of thang cause as we know the rescue cheerleaders show up for most providers and say how wonderful everything is for them.

But, the provider killed my mom.  "OH NO, he's a saint or waiting for sainthood".

 

[MannDude]

Understood. We've got stuff with CC in Chicago, I assumed if it was with us you'd notify me privately.

No need to name and shame if you don't want, completely understood.

Either way, those speeds are horrible. Have you requested to be transfered to a new node to see if this helps?

 

[drmike]

Yeah, the name shaming I am more reserved about these days.   I only bring public long after a provider has blown it far too long.

Last time I blew my top and asked for node transfer is when it suddenly was fixed.  When I signed up for service they were to provision at CC location but put the account in an entirely different datacenter and that required complaining and transfer.  Was a very awkward interaction on the transfer as they played stupid as if nothing was provisioned wrong. 

Almost the stereotypical VPS reseller experience one would expect with the Chicago folks and their tentacle nests around the country.

 

[Mitchell]

Sounds like they provisioned you in an extreme budget datacenter and now throttle you after they migrated you to their CC location to make you want to go back.


Just a random thought.

 

[Jack]

Location is Chicago.

Issue with the container is non stop.  Been this way for several months, minus the one time the provider actually did something and made it work right for < 1 week.

Iniz?

 

[AnthonySmith]

Sounds like the experience I had with 2host years ago 10TB bandwidth but dont you dare try to use it!

 

[Enterprisevpssolutions]

What kind of virtual tech are they using? Have they check the nodes to see if they are oversubscribed as you said the network is not an issues so we rule that out but other resources as disk i/o or not enough cpu and memory available on the node.