Snake needs to go in the garbage can....

[Frank Broughton]

FDSK for two days hmmmmm.... It would have been in my garbage can by now. Les, what manufacture drives you buying these days? The ones in Snake have been junk since day one. This is the third major outage in its not too old life.

It is so embarrassing to tell all the users of my VPS that all is well to only find out it started all over again. I no longer hide the fact that it is hosted here, I just show them the maintenance thread to save embarrassment of saying the problem is fixed when it is not.

Not blaming any person, but is this current system wise? Les, are you making these servers too large?

[jaypeesmith]

I think you might be right on this one. The sad part was that I was sort of bragging on JaguarPC the other day. Today, I am glad no one asked me about it. I would have been embarrassed. Looking over the history, there seem to be frequent disk issues with this node. I think it's about time for either a replacement of the node or, at least, the offer to move folks off of this node. I have some projects that I am hoping to deploy soon but, I don't feel comfortable right now because uptime is a concern.

[ShenPixel]

I've been questioning the practices in place for a while now myself. The long FSCK checks, they are just getting longer and longer each time there is an issue. But when I write about my concerns I feel like the downtime is being played down, like "the server was up just not accessible" type of deal. Les hes responded to me personally but I do not feel comfortable with his answers or the outcome. I do not see where Snake has improved (although when it came back online last evening page loads were fast for shot term), nor do I see suggestions I made being used.

Just how long is it acceptable to allow a FSCK to run, or allow a node to be down before you move your clients. In my opinion, JaguarPC needs to work on this policy the downtimes are WAY too long. If I personally owned this server, it would have been replaced some time ago and all my clients would have been moved to some other equipment/data center but I do not own it.

I was patient yesterday didn't submit a ticket until around noon, it didn't get answered for three hours, asking for patience. I said patience was granted but Jaguar was testing it. I also pointed out that the way to kill a "snake" was to cut it's head off and wondered if that has happened to this "snake".

Then I get up this morning and find it's down again, NO email in my inbox warning me it was going to be taken offline. I again wrote support and hours later no update, no update in the forum either since 4:30. All I can say to that is two things.

1. Way to go Jaguar PC, you really value your customers!! Thanks for letting us know snake was going to be taken offline. I know I sit on this forum 24 hours a day waiting for your posts that come every three/four or more hours when there are problems!! (Hope you can read into my sarcasm)

2. If TPTB cannot offer an amicable solution today or the latest Monday Morning, I am taking my business elsewhere! I lost two clients because of the past two extremely long down times. I reason I will loose at least another two after this weekend.

I'm not trying to rag on JaguarPC, I switched because of the results from a test account two years ago but the reliability hasn't been seen since I upgraded some time ago.

[Arshida]

More than 8hrs passed from the last update of the maintenance thread. We can do nothing but answering the customer's complaint calls and refreshing the maintenance page.... and WAIT.

[Frank Broughton]

More than 8hrs passed from the last update of the maintenance thread. We can do nothing but answering the customer's complaint calls and refreshing the maintenance page.... and WAIT.

I was just going to post that same thing. Natta since 5:30am. Just adds to or creates frustration not knowing what is going on. Especially when FSDK is not a solution for an existing problem - bad hd(s) due to bad power or poor manufacturing. I hope Les can tell us they mix up batches of hard disks and not use ones from the same purchase. Sure seems snake got a bad batch of them. I am just so used to doing things myself, relying on who knows who, is not fun for me.

[Frank Broughton]

Snakes history:

3-5-2011 - down "went read only" "bad hd replaced"
Maintenance - snake.nocdirect.com

3-6-2011 - down again FSCKing for 14 hours so far .....
Maintenance - snake.nocdirect.com

1-18-2011 down "FSCKing because of power outage" (perhaps this is how the other drives got damaged):
snake.nocdirect.com

11-14-2010 - down "Facing some issues" :
Maintenance -- Snake.nocdirect.com

7-23-2010 - down "snake down NIC card , other network issues" :
Maintenance - Snake.nocdirect.com

3-31-2010 - "down and is FSCK'ing" :
Maintenance - Snake.nocdirect.com

1-22-2010 - "down and our data center techs are working on it" :
Maintenance -- Snake.nocdirect.com

1-12-2010 - down "some memory related issues"
Maintenance -- Snake.nocdirect.com

[jaypeesmith]

I put in a ticket around 10:35am EDT about this outage. Around 12:08pm EDT, I received the standard, "we are sorry for the inconvenience" response and was informed that they are still running FSCK. I am concerned because, as has been stated here, there have been no updates to this issue in roughly the last 10 hours and, thus, no target time for resolution. I am just thankful that I am not currently dealing with any paying customers on this account. However, like I said, I am still concerned about launching any new projects from this node. When this is resolved, I will ask to be moved to another (and hopefully more stable) node.

[Frank Broughton]

The list above and also the dreaded openvz network outage bug that daily effected this node (taking it down, as far as the world knew, for short periods of time daily) for three plus months or so.

Nick, you ever get an answer for that bug from the openvz boys? I never did go back to that thread on the openvz forum.

So Les, is it time to purchase a new machine for snake? If so, make it MUCH smaller in size PLEASE, 14 hour FSCKs is killing us. What will happen when we fill this bad bear up? I am only using 23% of allocated space.

[ShenPixel]

still nothing?!? I went to town and searched for ricotta salta and support hasn't updated us?? Geez....I thought for sure that we would have some news after my lengthy search lol At least I was successful in finding the ricotta.

I'm curious as to how many clients are on snake.....I'm always one of the last to get back up because my domain starts with a w, anyone got an idea how many clients JaguarPC puts on a node/server and what is typical in the industry?

[eck79]

My VPS is on snake as well and have been very frustrated, but tolerant. Once Snake came back up yesterday our VPS (#102) would not boot. And of course there's no way to tell why (that I know of). So opened a ticket and that's when they took snake down a second time to FSCK.

Time to either fix this problem for good or we're off to another company.

[Box5]

My clients and I are gravely concerned about this server being down this long, and the lack of communication from JaguarPC. I suggested to my partners last year that we move to JaguarPC. I now regret that descision.

I will give this a few more hours while I search for a new host service.

[jaypeesmith]

I sent a response to ticket over 2.5 hours ago, asking if they had an E.T.A. for resolution. The response: **crickets**. I haven't heard anything back so, it looks like they are well on their way to going into a third day on this issue.

[jaypeesmith]

It looks like an update to the maintenance thread was made about an hour ago:

Maintenance - snake.nocdirect.com

It reads:

"Hi,

We are still working on this. The fsck is still running.

I'm looking into having alternate hardware setup for a restore just in case we can't get it back up. "

Almost 11 hours after your last update and you are just now "looking" into alternate hardware? How long is long enough before you actually deploy a new node?

[jaypeesmith]

It looks like a response came in around 1:20 ago: Maintenance - snake.nocdirect.com

My question is, in the 11 hours since the previous update, they are just now "looking" into deploying new hardware?

[JPC-NickO]

Hi jaypeesmith,

Your not getting a response because half the clients on the server are demanding updates which is why we have the network status thread. Level 1 technicians cannot do anything about the problem as they aren't working on it. They won't escalate it to level 2 because all I can say is what is in the network status thread. I try to update on the hour but sometimes I'm just to busy to spare a few minutes to say nothing has changed.

There is no ETA. These issues do not provide any real way to reliably predict when it's done and I won't guess.
I do not provide updates in tickets regarding these issues because it takes too much time away from working on this server and my other level 2 responsibilities.