Katarn slow?

[hmobrien]

Anyone else notice slow response on Katarn the last 24-48 hours? It's very noticable while using ssh. Everything was very fast previously.

-Mike

[hmobrien]

Still have problems on Katarn. I put a support ticket in and they said they didn't see anything that would cause performance or lag issues. Page loads are taking 5-6 seconds. I added a database and it took a full 45 seconds before it completed. Here's the output for uptime about 15 minutes ago:

23:17:40 up 1 day, 13:01, 2 users, load average: 5.03, 4.28, 4.08

I must be on a bad server or something is seriously wrong with my VPS.

-Mike

[ahodes1]

Katarn is totally botched. This is the second night that page loads are slow, and everything else was timing out.

[ahodes1]

Oh I love it... cPanel keeps paging me saying various things have failed and been restarted but I can't even get into a shell.. I got as far as it saying that I last logged in on May 11th (which is wrong), and then it died again. I can't reboot my VPS because VZPP is dead as a brick, support takes ages to respond, and I've got angry customers. I don't blame them, I am too. What happened to this great service I had for the first couple weeks?

[sysnop]

I'm on katarn.nocdirect.com. It's probably not a coincidence that right now I can't access my cp. Same problem that happened last night. Every night it's a slug, then something happens.

Not impressed.

[che1959]

There is a disk problem I suppose.

Try: df -h from the command line. I did earlier today, at 12:12AM CST and this is what I got:
vps:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 15G 708M 0 100% /
tmpfs 3.7G 4.0K 3.7G 1% /dev/shm

After I put in a ticket, things improved in about 45 minutes, but right now:
vps:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 15G 713M 4.0G 16% /
tmpfs 3.7G 4.0K 3.7G 1% /dev/shm

Basically, when my site went down it said I had 0 space available on the drive. Now it says I have 4.0G when I really should have closer to 14.3G available.

This same thing happened yesterday at about 12:00 AM CST. So my guess is that they are getting to the bottom of it. I just hope it doesn't repeat for a third day.

[rainboy]

Hmmm maybe you want to look for logfiles filling up your filesystem real fast ? Or is this not a VPS?

[che1959]

and right now:

vps:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 15G 717M 172M 81% /
tmpfs 3.7G 4.0K 3.7G 1% /dev/shm

At the rate it is going, sometime in the next 10 to 15 minutes, I am willing to guess that my server will be down to 0 MB available and that means no MYSQL server, with dynamic sites that means no sites.

Let's hope that doesn't happen, I am sure they are working on things.

[che1959]

It is not my logfiles filling up the system. It is the node having problems.

I have a 15gb partition, I am using 717 mb of it. I have only 172 mb available. My simple math says: 15GB - 717MB is 14.2 gb. 14.2gb is not equal to 172mb. Somewhere in the ether is a missing 14gb. The reply to the ticket last night said that the node was having a disk problem. Now today it is the same thing.

Now:
vps:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 15G 727M 0 100% /
tmpfs 3.7G 4.0K 3.7G 1% /dev/shm
vps:~#

And guess what? My sites are inaccessible.

[rainboy]

Ahrm.. sounds still like logfiles filling up a filesystem.. maybe not on your node, maybe on the hardware node, I do not have enough information about Virtuozzo to be sure that hardware node diskspace and vps-node are 100% devided, but that would be only logic to me, leaving your VPS OS to fill up the filesystem.

But guess if support is looking at it, they will fix it one way or the other.

Kindest regards,
Patrick

[rainboy]

okay i now see what you mean... this is indeed strange, and must be hardware node related, guess they only set quota and not make sepperate filesystems for each VPS node. hmmm wondering why not.. .that would atleast make sure this can not happen.

[che1959]

Yeah, actually it sounds like a program gone amok, maybe on the Virtuozzo side of the thing. It happened at the same time yesterday as it did today. Anyways, I hope it gets fixed and gets fixed for good. The first two week on the VPS were perfect and without a hitch.

My sites are accessible again as I now:
vps:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 15G 728M 459M 62% /
tmpfs 3.7G 4.0K 3.7G 1% /dev/shm

Don't know how long it will hold.

[floridaplowboy]

Mine did that last night also, but as you can see:

Filesystem Size Used Avail Use% Mounted on
/dev/vzfs 20G 2.6G 14G 17% /
/tmp 20G 2.6G 14G 17% /var/tmp

looks better now.

Hmm.... I do know that they do nightly backups, maybe the backup program
they use gone astray?

[rainboy]

or something silly as someone doing a grep on a directory and output it to a file in the same directory as you are grepping on ...

[sysnop]

It's so much better now.