amuck-landowner

BuyVM OpenVZ 128MB (NY)

Francisco

Company Lube
Verified Provider
Numproc is capped as a security net for our nodes. We felt 500 processes/threads was a very reasonable number and very rarely does anyone hit it.

It's better that than someone getting forkbombed and we're having to fight to get a node back under control.

Francisco
 

wlanboy

Content Contributer
Time to update my review.

What services are running?

  • Ruby scripts
  • Thin cluster
  • Rails app
  • lighttpd with static pages
Support:

Not a single support ticket needed.

Overall experience:

I really enjoy my vps in Buffalo. No hassles, no downtimes, no support needed.

The routings to europe could be way better but they are good enough for webhosting. I prefer rock solid hosting to lightning routing if I only have to care about http traffic.

I am using this vps for ruby based webhosting. The type of webhosting were your friends/family/sports club call you at 5 a.m. because NodePing sent them an email.

I did not get any calls yet - so everything is working as it should.

Yes it is OpenVZ - but not oversold:

Code:
free -m
             total       used       free     shared    buffers     cached
Mem:           128        114         13          0          0         61
-/+ buffers/cache:         52         75
Swap:          128         17        110
 

wlanboy

Content Contributer
Want to add the current status report:

statusbuyvm.jpg

3 hours and 14 minutes of downtime since March the 31st.
 

wlanboy

Content Contributer
A small update after the move:

statusbuyvmny.jpg

As you might know the 13 hours and 44 minutes of downtime can be ignored ;)
 

wlanboy

Content Contributer
Time to update the stats:

statusbuyvmny2.jpg

Not a single blip this year.

I/O is good, CPU too but the network throughput could be better.

To be fair - I only feel that network cap when I sync my backups from the vps to the EU located backup space.
 

Francisco

Company Lube
Verified Provider
Time to update the stats:

statusbuyvmny2.jpg

Not a single blip this year.

I/O is good, CPU too but the network throughput could be better.

To be fair - I only feel that network cap when I sync my backups from the vps to the EU located backup space.
Pull from our test file and see if it's doing the same thing at all. It could just be a poor route but if the speedtest (speedtest.ny.buyvm.net) shows better speeds then ticket :)

Choopa has been a real improvement all around.

Thanks!

Francisco
 

wlanboy

Content Contributer
Pull from our test file and see if it's doing the same thing at all. It could just be a poor route but if the speedtest (speedtest.ny.buyvm.net) shows better speeds then ticket :)


Choopa has been a real improvement all around.


Thanks!


Francisco
Might be the common problem of scp.

You speedtest  : 1.4 MB/s

My vps with scp: 850 kbit/s [nj-node09]
 

wlanboy

Content Contributer
Time for an update:

statusbuyvmny4.jpg

So after 102 days without any problems 2 hours and 44 minutes of network downtime occured.

Fran tweeted about the DC problems - so I did not need to write any tickets.

Hopefully Choopa gets this sorted.
 

HalfEatenPie

The Irrational One
Retired Staff
Howdy folks!  

Just letting you all know, I split the previous discussion about the frequency and updating of reviews  because?do=embed' frameborder='0' data-embedContent> it deviated away from the actual review content!  

If you'd like to continue that discussion, then please head over that way!

Thanks!  

Link (if the one above is broken): 
 
Last edited by a moderator:

wlanboy

Content Contributer
Time for an update:

statusbuyvmny5.jpg

39 minutes of network downtime since the last review.

CPU and I/O are good.

Network is getting better through the last weeks.


--2014-06-15 18:01:50-- 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'

100%[===========================================================================================>] 104,857,600 31.4M/s in 3.2s

2014-06-15 18:01:54 (31.4 MB/s) - `/dev/null' saved [104857600/104857600]

Uptime of the vps itself is 217 days.

Hopefully Choopa is able to get their network right (for more than one month) - they are on a good way.
 

wlanboy

Content Contributer
Time for an update:

statusbuyvmny6.jpg

1 hour and 6 minutes of downtime since the last update.

Uptime of the vps itself is 26 days.

CPU and I/O are ok.

Network is great.

Code:
wget cachefly.cachefly.net/100mb.test -O /dev/null
--2014-07-18 21:28:14--  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'

100%[===========================================================================================>] 104,857,600 41.7M/s   in 2.4s

2014-07-18 21:28:17 (41.7 MB/s) - `/dev/null' saved [104857600/104857600]
 

wlanboy

Content Contributer
Time for an update:

statusbuyvmny7.jpg

2 hours and 19 minutes of downtime since the last update.

Uptime of the vps itself is 43 days.

CPU and I/O are ok.

Network is great.

Code:
wget cachefly.cachefly.net/100mb.test -O /dev/null
--2014-09-28 10:43:56--  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'

100%[===========================================================================================>] 104,857,600 52.3M/s   in 1.9s

2014-09-28 10:43:58 (52.3 MB/s) - `/dev/null' saved [104857600/104857600]
 

wlanboy

Content Contributer
Time for an update:

statusbuyvmny8.jpg

2 hours and 47 minutes of downtime since the last update.

Uptime of the vps itself is 8 days.

CPU and I/O are ok.

Network is great.

Code:
wget cachefly.cachefly.net/100mb.test -O /dev/null
--2014-12-07 18:06:53--  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'

100%[===================================================================================>] 104,857,600 42.1M/s   in 2.4s

2014-12-07 18:06:56 (42.1 MB/s) - `/dev/null' saved [104857600/104857600]
 
Last edited by a moderator:

wlanboy

Content Contributer
Time for an update:

statusbuyvmny9.jpg

2 hours and 26 minutes of downtime since the last update.

Uptime of the vps itself is 19 days.

CPU and I/O are ok.

Network is great:

Code:
 wget cachefly.cachefly.net/100mb.test -O /dev/null
--2015-01-19 18:22:00--  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'

100%[======================>] 104,857,600 72.6M/s   in 1.4s

2015-01-19 18:22:01 (72.6 MB/s) - `/dev/null' saved [104857600/104857600]
 

wlanboy

Content Contributer
Time for an update:

statusbuyvmny10.jpg

1 hours and 11 minutes of downtime since the last update.

Uptime of the vps itself is 6 days.

CPU could be better and I/O is ok.

Network is great:

Code:
wget cachefly.cachefly.net/100mb.test -O /dev/null
--2015-02-15 22:25:06--  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'

100%[========================================================>] 104,857,600 49.5M/s   in 2.0s

2015-02-15 22:25:08 (49.5 MB/s) - `/dev/null' saved [104857600/104857600]
 

Aldryic C'boas

The Pony
Performance on NJ OVZ is horrible, to be honest.  The NJ builds are identical to the original LV deployment, which was before we started implementing SSD cache properly.  While none of the node's resources are actually oversold, the hardware is just having trouble keeping up (especially during common CRON times) now that the deployment is close to capacity.

We do have some major hardware upgrades going out to NJ within the next couple weeks to address these issues and bring the performance back to ideal levels.  I'll let @Francisco chime in on that one with more info, since he'll be doing the actual buildouts.
 

Francisco

Company Lube
Verified Provider
The NJ caches have been slowly starting to fail so we've had to kick them out which has been causing some of the IO spikes. If there was a long period of down it's likely a kernel panic + forced FSCK needed. It's also quite possible it was a leak on staminus that they didn't deal with. I know there was one that took a bit for them to cleanup. We do local filtering as well but sometimes the floods are just nasty.

The E5 nodes don't have CPU shortage but some of the 128's do. Originally the L5520's were just fine for them but things like buggy rsyslogd's, lots of munin's, etc, are taxing them at peak points.

With that being said, we already got some of the equipment in NJ for the big facelift it's due. The 256MB+ plans are finally going pure SSD's, the 128MB's are getting move up to L56XX processors, & new SSD caches for those too. We'll also be bumping the RAID cache from 128MB -> 512MB with newer cards. At this point we're just awaiting a shipment of RAM to LV before I head down to LV to personally prep the gear before it gets shipped to NJ.

Francisco
 

wlanboy

Content Contributer
Time for an update:

statusbuyvmny11.jpg

57 minutes of downtime since the last update.

Uptime of the vps itself is 7 days.

CPU could be better and I/O is ok.

Network is great:

Code:
wget cachefly.cachefly.net/100mb.test -O /dev/null
--2015-03-22 14:30:38--  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'

100%[========================================================>] 104,857,600 64.5M/s   in 1.6s

2015-03-22 14:30:40 (64.5 MB/s) - `/dev/null' saved [104857600/104857600]
 
Top
amuck-landowner