Wired Tree

Portal Home > Knowledgebase > Industry Announcements > Web Hosting Main Forums > Providers and Network Outages and Updates > Wired Tree

Posted by NeonRider, 09-07-2009, 03:07 AM
Could someone advise if it is normal for a hardware node to be rebooted twice in 12 hours period which takes several hours at a time? During that reboot sites and email are down completely. The rebooting is being done without prior notification and I suppose it is not possible to notify a customer in advance? It does not happen often and I am mostly satisfied with the service, but lately with a lot of improvements and free extra upgrade in server space the uptime has somehow deteriorated.
Posted by WiredTree Zac, 09-07-2009, 04:07 AM
No, it isn't normal, but I am sure you probably know that already. The server you are on is having hardware issues and our technicians are working to resolve them as fast as possible and get it stable again. When a server has hardware issues it isn't possible to notify you in advance since there is no way for us to know in advance. I apologize for any inconvenience this may have caused.
Posted by NeonRider, 09-07-2009, 11:35 AM
Thanks for letting me know about that. Last night it went down around midnight and recovered around 11 AM this morning. I guess there is no perfect server. I remember we were moved a lot and improvements and free complimentary extra upgrades were made, which is great. I suppose server hardware is very finicky and the cost of VPS is high, especially with this economical situation. So every downtime = loss makes one a little bit nervous. It's a bit like being laid off for a day or two. I hope the hardware industry will create perfect servers in the near future :-) Best wishes.
Posted by NeonRider, 09-07-2009, 02:19 PM
I just came back to the office to continue my work ad found that all my sites and my email and the WiredTree Grove are down again. This is a link to the image: http://img156.imageshack.us/img156/4...edtreedown.jpg . Wasn't going to post on here, but since Grove @ WiredTree is inaccessible at this time I was unable to post a ticket, so I'm asking on this forum for advise on how much more downtime will I have to deal with since I have urgent work to do and please understand the income is down very much with this weak economy, so any downtime hurts 10 times more than before. Please advise if there is WiredTree's private forum to post inquiries about downtime when WiredTree.com is inaccessible, if you don't want this in public. Also a question: perhaps I'm missing something, but shouldn't host's support website be up all the time? Thank you in advance for your advice and responses.
Posted by 1boss1, 09-07-2009, 03:01 PM
We must be on the same host node, all down here also.

It's been running horrible for a while, for the last 21 days Pingdom at 1 minute monitoring shows:

Quote:

54 Downtimes
Downtime: 17h 10min 12sec
The longest downtime was 6h 3m 58s on 07/09/2009

Hopefully it will be fixed soon.
Posted by NeonRider, 09-07-2009, 03:17 PM
Wouldn't it be a good idea for a hosting company to insure their customers from hardware/software failure resulting in a significant downtime? What happens if customer's site/server is down for many hours, an entire day or for several days? Who absorbs the loss and who is responsible for the damage? God? The manufacturer? Is it fair to blame hardware for the failure since hardware is not a human and not a business and can't reimburse the customer for the loss. Should apologies be enough when a website downtime occurs? Technicians are working, I understand, then they fail to fix it properly and a downtime occurs again, and again. Then what? Usually when a company fails, they apologise, but apologies should not be enough if there's a continued, repeated failure, a customer should be reimbursed for the loss in some way. Not that I'm asking for anything. When my business fails or makes a mistake, I, instead of apologising refund part or all of the customer's payment and/or give a good discount for the next order, add bonus material etc. Shouldn't customers websites be transferred to another server if there's a bad failure on their server while that server is being repaired? Why does customer have to wait for hours and days while the server is being fixed? Shouldn't this issue be covered under insurance, so when downtime occurs the customer is paid for the loss by the insurance agreement? All these questions, I hope someone can answer and comment them while I'm waiting here, losing a second business day to downtime. To me "Labour Day" means hard work, no rest. I'm glad Wired Tree works today. Some hosts I hosted at were off on weekends :-)

P.S. I just checked at 3:26 PM and I'm up again. So they fixed it pretty fast this time, in about 1 and 1/2 hour. Hopefully it will not go down anytime soon. :-)
Posted by WiredTree Zac, 09-07-2009, 03:24 PM
Grove is up and running - it hasn't gone down at all so I am not sure why you cannot connect. It gets checked on 1 minute intervals from outside sources and no issues have been reported. The hardware node you are on crashed again. It is already back online and VPS have been starting, but we will need to take it down to swap hardware later today to permanently address the issue. Thank you for your patience and I apologize for the inconvenience.
Posted by 1boss1, 09-07-2009, 03:29 PM
I'm not sure what's going on, but the node has been horrible for a couple of weeks.

This isn't normal, i've been with Wiredtree for "almost" 2 years and to their credit this is very uncharacteristic of them. Not much consolation when everything's down i know. I've only opened 1 ticket since May until yesterday.

They had scheduled downtime the other week to fix a hardware problem on my node, and since then it's been heaps worse so it seems whatever was done didn't work. Actually uptime for the month so far has dipped in to the 80% range.

I can get in to Grove ok, so try again if you can't.

Edit: Zac posted as i was typing, yes i've had no issues loading Grove.. But so far my VPS is still down. Hopefully it will come to life soon.
Posted by WiredTree Zac, 09-07-2009, 03:35 PM
We knew about the performance issue which is why we scheduled maintenance to upgrade the server. Some component of the upgrade (even though everything was tested beforehand) has caused the server to be unstable which our technicians have been working on debugging and fixing. We will be swapping hardware in unison with moving VPS off onto different servers. The server had over 400 days of uptime at one point before we had to reboot it for a kernel update and we are going to do everything we can to get it back to that level. Please be patient. This is a high priority issue, but issues like this do take some time, and obviously if the server is down we cannot move anyone off of it.
Posted by 1boss1, 09-07-2009, 03:47 PM
I just threw it in the calculator and i've been with Wiredtree for 616 days.. I'm pretty sure i've been on the same node and yeah it's been good until the last few weeks.

Hope she comes back up soon, and the new hardware makes it snappy again. Even though i only average ~300mb RAM from 1GB and little CPU it's been really slow when it's up.

Thanks for the update Zac.
Posted by 1boss1, 09-07-2009, 06:15 PM
Is anyone else still down... Or is it just me?
Posted by WiredTree Zac, 09-07-2009, 06:25 PM
We sent out an email about an hour ago saying that we will be taking the hardware node down momentarily for a chassis swap. This has just occurred ahead of schedule and everything is coming back online. Before that all of the VPS had been back online for awhile (the amount of time varied). I am showing all VPS started as of right now but it will take a few minutes before everything is fully started and back up to full speed. If you are still having issues in about 15 minutes, please open a ticket and we can check. We will of course be monitoring it closely.
Posted by 1boss1, 09-07-2009, 06:42 PM
Ok cool thanks Zac, i didn't get an email and also ~24 hours ago when i submitted a ticket about the downtime i didn't get a confirmation email to my ticket like normal.

I have my ISP email, plus one attached to my server on file so it must just be sending to my down account and not both.

Pingdom is starting to show green lights now, looks like we are in business.

Thanks for the updates, fingers crossed this new chassis is kind to us.

Edit: I have Gmail + ISP email on file now, i dare it to route my mail down the wrong intertube now lol.