Major Reliability Issues

Update. The "Hardware Issue" fix didn't resolve the issue, so after losing time already we have to migrate. Now, after 40 minutes, the migration is only at 2.2 percent…so it looks like it will take about 28 hours to complete. Support says the slow throughput is likely because of the hardware issue on the server we're migrating away from.

So…our website down for 28+ hours because of additional Linode hardware issues? Definitely not acceptable.

For everyone else's benefit I truly hope that we are alone in our experiences.

EDIT: I'm not here to bash Linode. Yes, we are extremely frustrated by recent events. But the main purpose for this post is to see if others are experiencing issues…to help us decide whether to stay with Linode or go elsewhere.

I've had no issues on 5 Linodes (in 5 of the 6 datacenters) that I either use or help administer.

@markb1439:

Ever since the new SSD architecture, we've had tons of trouble. Are we alone or is anyone else experiencing repeated issues?

In just a few weeks, we've had trouble backing up and restoring. Then there was a hardware issue requiring a migration. Immediately, there was an issue on the host we migrated to. And tonight our website has been down due to what turned out to be a "Hardware Issue" on the host where our Linode is located.

Linode used to be rock solid but we've had problem after problem ever since the new architecture came out. Are we just "lucky" or are our experiences the new norm?

Thanks!

My Linodes have been running fine and fast. I use rsync to back my hosts up to an off site location. Works great.

Thanks to both of you for sharing your experiences. I am optimistic that your experiences are typical. I'm just gun shy because this Linode is on its fourth host in a few weeks (three migrations due to hardware issues).

In fairness, I should point out that another Linode has been rock solid for the same time period.

And Linode support is very responsive and helpful.

I manage a lot of nodes and since migrating to the new nodes I've not seen a change in the number of hardware related inccidents.

After getting enough data to predict a 28 hr migration, why didn't you just spin up a new node, config, restore data - something that should be way quicker then 28hrs (although I guess that depends on how much data you actually have).

> After getting enough data to predict a 28 hr migration, why didn't you just spin up a new node, config, restore data - something that should be way quicker then 28hrs (although I guess that depends on how much data you actually have).

Good call! That's exactly what I did, spun up a new Linode, did an IP swap, and restored the daily backup to the new node. So no, I didn't end up having a day of downtime, but I was still frustrated about yet another hardware issue and the need to move to another host again.

I can confirm that I've been having similar issues. i've had to migrate my DB server 3 times, and there were times when I haven't changed any config, servers been running fine for months, and without any server upgrades or changes or restarts, and then suddenly, just spontaneously the servers would experience outage, and would not work until the DB server was migrated again.

The staff are super helpful but even they can't figure out what the issues were. And then for no reason, the website will start to function properly again.

@cloudry, thank you for sharing your experiences. At least I know I'm not crazy.

Just curious…which datacenter(s) do you use? My issues are in Dallas. The Linode in Newark seems fine.

Hi,

I use the Dallas data center, have three nodes one of which is a 1024, the others being 2048. I have only experienced solid stability on all three nodes, no down time for me yet unless I do something that would cause it, like a reboot or resize of the node.

Blake

@Tech10, thank you for sharing your experiences too.

@cloudry, just curious, what size Linode is that DB server (if you don't mind sharing)? The Linode I've had so much trouble with is a 16 GB. The one in Newark that is doing fine is an 8 GB. Another is a 1 GB, which is also doing fine. I heard that each host only hosts the same size Linode, so I don't know if that has anything to do with it.

I have only been with Linode for a little over a month but my Ubuntu 14.04LTS 1GB Linode in Newark has been up and running my LAMP server with absolutely no downtime at all…frankly I'm amazed at the speed and reliability of Linode.

You know I spoke too soon, I've had two in a week now! Murphy stalks me :(

Had 1 go down now. Ticket says it is resolved when it isn't and there has not been any updates. :(

You might want to re-open that ticket, or open another one. If it's marked as resolved, you probably aren't going to get any more updates on it.

And now it's down again. This is ridiculous.

markb1439 - would you please PM me, or post here, something that I can use to find your account? I'd like to investigate the situation further.

Thanks,

-Chris

I have been having issues but it is on a pre-SSD server. I have had two sites that have been what seems to be hacked. I have had to restore to an old snapshot since Apache refused to start even after quite a bit of trouble shooting. I have changed all passwords and tried to make them much more secure.

I don't think my issue is a Linode issue. A third party software vendor was hacked recently and they may have retrieved some of my information from them.

Running a server is a challenge for me. If it were easy, I probably wouldn't bother.

fos

@markb1439:

@cloudry, thank you for sharing your experiences. At least I know I'm not crazy.

Just curious…which datacenter(s) do you use? My issues are in Dallas. The Linode in Newark seems fine.

We have always been using the Dallas datacenter and things have been fine.

I migrated to Dallas a while back when Fremont was the most unreliable DC in the world. Except for a minor network issue earlier this week the service has been totally stable.