Recently I was in the market for a managed server so of course I did my research here at WHT. After spending a few hours digging through old posts it became evident that ServInt was at the top of my short list. Well it has been less than a week but I can tell you that I have been thoroughly impressed by the support that ServInt provides.

I just signup their service today and waiting account activation (almost 12 hours now) :(

12 HOURS!!

That is forever in the web hosting world.
I say:

sack em'

:cool:

I got my VPS account with ServInt Took some waiting to get it though


They notice problems on the VPS Server before i do, http goes down, they know about it and restart it for me. Great service

Servint is good, but when you have a bigger than normal issue then get ready to have poor support.

FOr example, if your website HTTP goes down, sure they will fix it immediately for you.

But say, your HD had a crash or something occured with your kernel.. then say hello to 8+hr downtimes, which would take another noc about 1-2hrs to fix would take Servint 8+hrs or more.

I'm sorry, but how can you justify 12 hours?:confused:

Originally posted by fusionx1
I'm sorry, but how can you justify 12 hours?:confused:

I'm sure ServInt gets numerous orders a day. Checks need to be made on each order (e.g. ServInt doesn't do cheap prices) so they will have to be extremely vigilant for fraud.

It seems reasonable to me.

Andrew

Originally posted by fusionx1
I'm sorry, but how can you justify 12 hours?:confused:
It has been stated many times that although we run a 24/7 NOC, we don't run a 24/7 sales/billing department. Because of repeated fraud attempts, we manually screen every order. His order will be processed shortly. I justify 12 hours because it means alot for the security and integrity of our clients--they certainly enjoy that once they're online, I hope you understand.

Originally posted by universal2001
Servint is good, but when you have a bigger than normal issue then get ready to have poor support.

FOr example, if your website HTTP goes down, sure they will fix it immediately for you.

But say, your HD had a crash or something occured with your kernel.. then say hello to 8+hr downtimes, which would take another noc about 1-2hrs to fix would take Servint 8+hrs or more.

I'd love to see you back this up. You were asked to in a previous thread, but never did so.

I never intended for this thread to start a bashing. I wanted to post a testimonial in the "sticky" sale that ServInt has but the mod I asked gave me the indication it would be best suited here in the main forum.

In regards to the 12 hours. I did have to wait about that long also. This is acceptable to me. I wouldn't want to wait that long for regular hosting but for a dedicated or VPS I don't mind. Instant gratification can be a bad thing sometimes!

Originally posted by inogenius
I'd love to see you back this up. You were asked to in a previous thread, but never did so.

I can back it up. I have several servers at Servint. If there are hard drive issues between 9 am and 5 pm EST they are handled promptly. If you hard drive dies after that you are SOL while you "wait for the engineer."

This is a chronological outline of events that happened when a drive failed on one of my servers at about 3 a.m.

I received an email from the support person at Servint acknowledging a problem with the drive at 4:36 a.m.:

"This server has been very problematic tonight and it appears as if there are drive problems. Our on call engineer is looking to recreate the primary hard drive first thing when he gets in this morning. We currently have a trouble ticket opened on this issue and will keep you updated with all events concerning the server."

At 7:00 a.m. I inquired about the status of the drive and if the engineer had started working on it. This was the response:

"I will notify you when he begins work on the server."

At 8:36 am. the engineer finally arrived and started work on the server:

"The engineer is working on your server now. We will let you know any information that we get."

At 9:52 a.m. I received this response from the noc:

"The engineer is currently reinstalling cPanel on the new drive. Once that has been completed he's going to try and restore the server from the cpbackup on the secondary drive."

At 9:55 a.m. I received another response:

The primary drive appears to have failed. It will not boot or be detected at all so it's not likely that we'll be able get much evidence as to what happened to it or why."

At 11:54 a.m. I asked for a status report:

"The engineer is still working on it."

At 12:39 p.m. I sent the following email to Reed Caldwell:

"Hi Reed,
I had a hard drive die this morning at 3:00 a.m. Even though your
engineer was called, he did not arrive and start working on the server until 8:44 a.m. It is now 12:39 p.m. EST and I am getting sparse updates from your data center and the websites on that server are still not up.

What is your hardware replacement policy?"


At 1:28 p.m I received an email from the Engineer stating that the server was back online. The engineer had restored the sites, but when he setup cPanel he did not use the hostname of the server (which was prominent in every ticket) and did not use our nameservers. Every site on the server was pointing to the wrong IP address.

At 3:11 p.m. Reed responded to my email:

"I'm on the road right now and don't have access to the NOCDB, but I'll do what I can. Which server was this? Was it the primary hard drive that the server boots from?

Sincerely,

Reed J. Caldwell
ServInt Internet Services
http://www.servint.net"

I have never heard from Reed again regarding this matter.

I have had a drive fail at 1 p.m. that was mirrored and replaced by 2 p.m.

I had a hard drive fail on a Saturday once that did not get fixed until Sunday, and that was only after I emailed Reed asking why it was taking so long to get the drive fixed.

Servint's support during working hours is among the best in this business. If you need anything other than a reboot after hours prepare for a long night.

Servint's engineers only work during office hours...so if any thing more advanced such as more than reboot/restart happens on non-office hours....You will wonder why you pay such a SWEET price to them..

How could we control HD NOT to fail on weekends? :D

Ouch, that's dissapointing after reading all the good reviews about them.

"This server has been very problematic tonight and it appears as if there are drive problems. This doesn't sound like the server was offline, but simply having problems. That would make the statement about the Engineer coming in to recreate a drive much less of an emergency. Was the server offline from the time the first ticket was submitted?

Looks like the perfect justification to use RAID along with a 3rd back-up drive to me.

Originally posted by Mark_TVI
This doesn't sound like the server was offline, but simply having problems. That would make the statement about the Engineer coming in to recreate a drive much less of an emergency. Was the server offline from the time the first ticket was submitted?

Looks like the perfect justification to use RAID along with a 3rd back-up drive to me.

The server was down, the drive was dead, which was confirmed by the engineer's response:

"The primary drive appears to have failed. It will not boot or be detected at all so it's not likely that we'll be able get much evidence as to what happened to it or why."

The tech tried to reboot the server several times but could not successfully bring the server up, thus his statement that "This server has been very problematic tonight and it appears as if there are drive problems."

Originally posted by Mark_TVI

Looks like the perfect justification to use RAID along with a 3rd back-up drive to me.

Mark,

I understand you like ServInt, and as a former servint customer, Reid does run a nice operation

You're trying to tell me the customer should spend extra money on RAID (Let's be real...ServInt isn't the cheapest, nor the most expensive) because the customer can't rely on servint's engineers to fix problems in a timely matter?

The customer shouldn't have to wait 12 hours for a hard drive replacement. Maybe there's more to the story, but you need to fall back into the real world where even servint can do wrong. :D

The server was down, the drive was dead, which was confirmed by the engineer's response The ticket was submitted at 4:36 AM yet the drive was not described as dead until 9:55 AM, that is why I asked if the server was indeed offline this whole time. There was no indication from anything you posted that the server was offline until 9:55 AM.

I certainly do agree that 12 hours is far too long to wait for a drive replacement (if that is exactly what happened, we've only heard one side of this issue so far). However I also am a firm believer in taking as many precautions as possible to avoid situations exactly like this.

Nickn, you certainly don't have to agree with our philosophy of trying to plan for situations such as this but it has always worked quite well for us.

I should also add that for a company like ServInt that has been around for over 9 years, this is the first time I've ever heard of anything like what Curious Too is claiming happened to him/her...

Originally posted by Mark_TVI

Nickn, you certainly don't have to agree with our philosophy of trying to plan for situations such as this but it has always worked quite well for us...

We use RAID, and have physical access to our machines.

However, there's still no excuse for a 12 hour wait on a bad hard drive. my guess is the sales pitch wasn't:

"Would you like RAID?"
"No thanks..."
"We're likely to take 12 hours to replace your drive, are you sure?"
"In that case, sign me up."

Nothing against ServInt, simply pointing out it's not the user's fault because they didn't get RAID. :)

I never said that it was the user's fault, I said this is the exact type of situation that is the justification for using RAID and a 3rd drive for back-up.

You must have also skipped over the part where I agreed that a 12 hour wait for a drive replacement was unacceptable.

Originally posted by Mark_TVI
The server was down, the drive was dead, which was confirmed by the engineer's response The ticket was submitted at 4:36 AM yet the drive was not described as dead until 9:55 AM, that is why I asked if the server was indeed offline this whole time. There was no indication from anything you posted that the server was offline until 9:55 AM.

Sorry for not making that clear. When the server went down I did not open a ticket, I phoned the noc instead. The tech was already aware of the problem with the server and was already trying to bring it back up. At 4:36 a.m. he sent me what was in effect a "status" email.

We had some major problems with ServInt. We've moved to a different provider.

Originally posted by SpiritAu
We had some major problems with ServInt. We've moved to a different provider.

Well, it seems they still work the same from when we left as well.

Mark_tvi, you guys sure are servint's best cheerleaders. :D

We can't seem to get quality hosting from US providers these days.

We’ve moved ~90% of our hosting capacity to Australia and haven’t been happier.

Originally posted by net-trend
Well, it seems they still work the same from when we left as well.

Mark_tvi, you guys sure are servint's best cheerleaders. :D Probably because any issues we ever had in more than a year that we've been with them were always resolved quickly and professionally. In fact this is the first issue involving ServInt that I've seen on these forums that had any time-line documented. That in itself is pretty amazing considering the number of clients ServInt has that post here.

Servint provides excellent support during working hours. The techs who provide support after hours do very well at what they are allowed/trained to do. There have been several times when I have logged into a server to check a problem only to find someone from Servint already logged into the server and performing diagnostics. But in my own personal experiences a distinct pattern has developed in regards to after hours and weekend support, and that is if a critical issue arises during those time periods that is beyond the capabilities of the techs that are on duty, the server will be down for an extended period of time.

lol

everytime Servint is running special, troops come up here :D

Originally posted by Curious Too
Servint provides excellent support during working hours. The techs who provide support after hours do very well at what they are allowed/trained to do. There have been several times when I have logged into a server to check a problem only to find someone from Servint already logged into the server and performing diagnostics. But in my own personal experiences a distinct pattern has developed in regards to after hours and weekend support, and that is if a critical issue arises during those time periods that is beyond the capabilities of the techs that are on duty, the server will be down for an extended period of time. The first portion of your post has been the same as our experiences. Fortunately we have only had one major crisis occur after hours and it was resolved within 30 minutes with only 2-3 minutes of downtime.

Hopefully we won't ever have to experience what you appear to have gone through Curious Too...

TVI 's analysis to Servint customers' problems has been often looknig like more detailed complete than official statments :D

Originally posted by universal2001
Servint is good, but when you have a bigger than normal issue then get ready to have poor support.

FOr example, if your website HTTP goes down, sure they will fix it immediately for you.

But say, your HD had a crash or something occured with your kernel.. then say hello to 8+hr downtimes, which would take another noc about 1-2hrs to fix would take Servint 8+hrs or more.

I have experience problem with kernel, and the technical support can fix only left than 1 hours, they directly recompile kernel from source. I'm forget the name, Giles or Jim.

Originally posted by SpiritAu
We had some major problems with ServInt. We've moved to a different provider.

Could youd share us your experince here ?

We had a PLESK VPS as I recently brought out a small hosting business that was running PLESK with about 80 domains on a reseller account. We moved from the reseller to PLESK VPS, as I can't live without root.

Everything was fine for about 20-30 days after the move, and then at about 1am I get an SMS from our monitor saying HTTP was down. I login and it won't start at all. In the end it wasn't anything to do with Apache but Virtuozzo wouldn't let apache bind itself to the loop back address.

Down 18+ hours.

Very little assistance/updates from Servint. They blamed PLESK/SW-Soft.

We moved to a dedicated server here in Australia and everything is working great - 100% uptime.

Originally posted by Jori
lol

everytime Servint is running special, troops come up here :D

hey junior, I was posting "my impression" which I did on my own. By saying "troops" you suggest that I received guidance from ServInt.

I would ask that you show me ONE provider on the forums that doesn't have anything negative posted about them anywhere. I have the utmost respect for the guy (he's a member here) that previously provided all my services and even he has a few negative posts.

Originally posted by SpiritAu
In the end it wasn't anything to do with Apache but Virtuozzo wouldn't let apache bind itself to the loop back address.


Hi,

You know, I would not expect any difference in binding to loopback address in Virtuozzo comparing to dedicated box. Could you tell exactly what was the problem and why you expect that it is Virtuozzo-related?

Thank you
--
Andrey

Originally posted by moruga@
Hi,

You know, I would not expect any difference in binding to loopback address in Virtuozzo comparing to dedicated box. Could you tell exactly what was the problem and why you expect that it is Virtuozzo-related?

Thank you
--
Andrey

They said it was, and the error logs kept showing can't bind to 127.0.0.1 and 0.0.0.0.

I brought that to there attention but go no reply.

If I still had the VPS i'm sure I could dig up the logs.

Originally posted by Curious Too
I can back it up. I have several servers at Servint. If there are hard drive issues between 9 am and 5 pm EST they are handled promptly. If you hard drive dies after that you are SOL while you "wait for the engineer."

I would really like to hear Servint's point of view on this.
I've only had one server issue on a weekend over a year ago.
I called the NOC (saturday night) and I was told I'd probably had to wait until Monday for the engineers to take a look at it.
Fortunately I managed to fix the issue myself that night :D

From reading the experiences of others here it seems the situation with 'after hours' support hasn't changed.

Servint support during office hours is the best I've ever experienced, but server problems can occur at any point of the day or night.

If having experienced admins at the NOC 24 hours a day is not feasible, then perhaps have someone on standby who is able to get to the NOC within 1 hour?

I'd imagine that as Servint grows, they should reach a point where they can afford to have experienced admins in the NOC 24/7.

Nevertheless, I'm still a Servint fan :)

Hello guys,

sorry for all your problems. month&half a ago, I had also bad hard drive, and it was during the weekend.

It was solved very fast, could not say anything.

---------------------
Date: Sat, 3 Jul 2004 08:11:06 -0400 (EDT)
From: Network Operations Center <noc@servint.com>
To: XXX
Subject: Re: bad hard drive
--------------------


N.

Originally posted by NEMON
Hello guys,

sorry for all your problems. month&half a ago, I had also bad hard drive, and it was during the weekend.

It was solved very fast, could not say anything.
N.

Very good to hear. :)

Originally posted by universal2001
But say, your HD had a crash or something occured with your kernel.. then say hello to 8+hr downtimes, which would take another noc about 1-2hrs to fix would take Servint 8+hrs or more.
1-2 hours to fix a HD failure? Not sure if that's just getting the new drive in there, or a full OS and account restore, but it's going to take much longer than 1-2 hours to get the server operational, following a HD replacement. Maybe 4 to 5 hours min.

Frankly, I think you should talk to Reed or Christian about how to prevent this issue from happening again in the future. I've had servers with ServInt for over 2 years and never had any HD failure. I don't use RAID, but I have full backup in the secondary HD using rsync and the backup script they provide.

I forgot when the last time I had emergency issues like kernel panic or serious server crash, but as far I remember, those were resolved quickly by their senior engineers (either Giles, Joe, or Jim).

I had been with Servint for 20 months now and I havent had a hard drive problem yet. I have 2 drives in RAID. Hope this could ease a hard drive failure when it happens to me :) Should I add a third backup drive?

Anyway, I only remember having a major problem with my server: bad RAM that was replaced very quicly, the server was even moved to a lab for testing. I think I just had a few downtime minutes and the issue was corrected.

So, support has been really great for me :) although it would be nice to hear Reed talk about this situations in order to prevent them. Right now i have many clients that start calling us when the server is unavailable for 10 minutes, I cant imagin what would happen if i get an 8 hour downtime. :(

universal2001, what was your issue with servint support? As far I remember you had been with them for two years more less. Actually this is the first time i read a not so positive comment about servint from you.

Best regards, and I still can say (imho) that servint rocks!

iago, as long as you do remote offsite backup regularly, you won't need 3rd backup drive :)

Originally posted by Curious Too
...Our on call engineer is looking to recreate the primary hard drive first thing when he gets in this morning...

Now I'm confused: define 'on call'?

Originally posted by BigBison
Now I'm confused: define 'on call'?

on call usually means 'when we manage to wake the guy up', see the related theplanet threads ;] we've recently run into this with a client's transport circuit and ilec technicians.

a couple thinking points:

* raid is *not* a replacement for a snapshot backup
* raid *is* meant to prevent extended downtime on a single failure
* regardless of raid (or lack thereof), drives *must* be replaced expeditiously, which 12 hours is not
* be glad you aren't with ev1

as usual, i will caution everyone to defer passing judgement until reid chimes in with his side of the story. quite a few people backed the statements up, but the picture is never complete when you see one half of it.

on a related note, some folks' "my provider can do no wrong" attitude is rather vexing. i'll mention peer1 and myself before you do and ask that you please review the appropriate threads where i call 'em as i see 'em even wrt our upstream.

paul