New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
https://status.hosthatch.com/
Unfortunately, that this is by location and not by node, so they won't say a location is offline for the puny little 5% that lose connection. Pfffft
This is only useful if an entire location goes down, practically useless IMO. Needs to be like how most providers do it, with status per node.
Yeah clearly that status page could use more details
Looks like ramnode are using 'Node Ping', while HostHatch are using 'Uptimerobot'
One might assume bringing these critical issues to the two head honchos would get some traction and some appearance of getting off their asses.
Nope
HostHatch will be the company that ruined two year prepayment deals for me. Any company I see that offers two year deals I'll think, "oh shit, they're looking to check out like HostHatch. Time to move on".
I have been unable to use LA for 5 days
Been down for a week or so - anyone else? VPS says it's running but won't boot...
Hosthatch deadpool incoming?
Uptime was good until now - no help from support, a VPS that won't boot. No status updates. Fuck you maybe?
Just assumed mine was up, I need to check, thanks for the heads up...
If you're just trying to connect using SSH, connect to VNC via HostHatch's control panel and see what it says? It's likely something is corrupted - either the kernel, the GRUB config, the fikesystem itself (fsck needed), etc. Post a screenshot of what it says.
you may have to go fsck yourself
"no setup signature found"
Trying to boot into old kernels seems to work but every command i issue gives me a segmentation fault.
Did anyone's VPS come back or was this a fake restore where literally every node is down and out?
Even if their entire LA location went poof, impacting both storage and non-storage VPSs, that doesn't mean they're deadpooling.
That's just 1 site out of 14?
"No help" and "no response" are two totally different things.
I'm expecting the latter to be the actual case - please correct me if I'm wrong.
Ok, but this has already been highlighted somewhere in this thread and clearly it's something they'll need to work on and improve.
I know you're frustrated, but sorry, cursing won't revive your VPS so maybe have some respect for other human beings, especially to the ones who could actually help you in this situation.
I have several services with them too and it's a known fact that their ticket response time is horrible so the best thing for you to do is either attempt to find out yourself what's going on, as suggested by @Daniel15, or just wait for them to respond to your ticket.
No guarantees that they'll respond here, or even want to, but I'll just tag them for the sake of helping.
@Emil @hosthatch
✌️
The segfault most likely means the program is corrupt and it's trying to execute invalid code or access an invalid memory location.
If it's Debian or Ubuntu, does
aptwork? If so, you can useapt --reinstall install whateverto reinstall the program, which should give you a working version. If apt is corrupt then I'd guess you'd need to boot from a live CD and copy the corruptedaptfiles across manually. Once apt is working, you can reinstall any damaged apps.I don't know much about CentOS or other distros, sorry!
In any case, your data is likely at least 1% corrupted (I found 1-5% corruption across all my data via checksum mismatches compared to backups), so I'd recommend restoring your latest backup. In my case I only back up the data and not the operating system, so I've also had to manually reinstall a bunch of programs.
Rule #2 of low end boxes is to always have backups. Borgbackup is efficient enough that you can run it daily - it'll likely only take a minute or two to run, as it only has to back up the differences since the last backup.
(rule #1 is "no more than $7/month")
They restored it inasmuch as the VPSes power on now. Previously they weren't even doing that. However, it seems like many of them are corrupted. As far as I'm aware they haven't attempted any data recovery on the corrupt data.
@hosthatch hey, there seems to be several complaints with no reply at all. Do you guys have any information you can share?
This is essentially what happened to me (though I couldn't even get it to boot). You can use your Linux distribution's package manager commands to verify system packages - most likely some libraries are corrupted. If you can't even run the package manager tools, boot from a rescue distribution, and then use tools like md5sum/sha1sum to compare checksums on important files and then copy them over. It's easiest to do this if you have another system (e.g. a VM on your computer) running the same distribution and package versions as the broken system, though in general downgrading libraries by a few minor versions won't cause any problems.
Once your package manager is working, you can then run something like dpkg --verify or rpm -Va, which will check your installed files against the checksum files. If any packages report errors, reinstall them. If your distribution's checksum files are missing/broken for any packages, reinstall those packages anyway.
Once the OS is back up and running, you'll probably have a similar issue with your data files, but unlike system packages there's not much you can do about those unless you have backups of the data elsewhere.
I'm not sure if Hosthatch will provide this level of support and reconstruction - I think other people in this thread took the DIY approach.
Honestly I'd be surprised if any unmanaged host provided this level of support. Some (including Hosthatch) might do it as a one-off courtesy, but the cost to a low-end host in terms of support hours would end up costing more than any revenue gained from the customer, so it's not really sustainable if a lot of people also ask for the same thing.
The idea with low-end hosting is that you're intentionally giving up some amount of support in exchange for cheaper hosting. This applies even moreso in HostHatch's case since their sale prices are a lot lower than many other hosts. For people not comfortable with the low-end-lifestyle, I'm sure the $50/m fully managed VPS providers would be happy to have them
I'm not defending HostHatch or saying their approach is good - they really should be more active in terms of messaging and replies on this thread. I'm just saying that it's somewhat expected to get less hand-holding and less support when things go wrong the lower in price you go, in general across all low end hosts.
At this point, @hosthatch should be in damage control mode and bring his ass here to "face the music". Right now, it's not looking good at all.
We've got people (including myself) considering other options when the 2-year renewal falls due, and this thread is proving to show a good case for letting it expire.
A quarter after 2020's BF I gave away my account with them because of having many small problems here and there but not getting a support reply (or just a useless one like "We're looking into it" without something that has changed in the end). Last BF they promised everything has improved so I gave them another try and it first appeared to be running better but now there's again a problem here and another one there (not storage line) and support is again non existent. This was my last try. Using their service is just a waste of time for me now even though it's cheap and if it's running, it's a good service. The latter one is timely limited unfortunately and support abilities to restore this state are too limited to make it any fun (in terms of usefulness) to use their services.
mine is still working
I am afraid we have different idea of what "slow support" and "hand-holding" mean.
But that's not the case with hosthatch company.
Weeks of downtime because of their broken system, broken control panel, and so on without a single support ticket response is not just slow support. It's pure ignorance, incompetence, incapability to manage own system, to manage own userbase, to provide them what the paid for.
Once THEY fuck up you're pretty much on your own. When they fuck op, not when you fuck up.
You may be able to put OS back online, able to reinstall system. Or you may not be able, is out of your control because of their broken control panel. It doesn't matter. You're own your own and you're still paying for it.
You may find yourself in a situation with hosthatch where control panel wont let you to reinstall operating sistem. You will open support ticket. And there won't be a single response for a several weeks. This means several weeks of downtime you're still paying it. And that's not slow support response, that's just rip off.
Some people says "for what I am paying they are good". I am wondering what those same people will say next time when they won't be able to use already paid but suddenly inaccessible VPS for a month or more and hosthatch won't give a shit about it. Literally won't give a shit, because they won't bother to respond even simple support ticket inquiry.
Considering that the context of your quote was someone discussing whether or not HH would help them rebuild their OS, I think your definitions line up pretty well. It definitely would've been hand holding. Glad we're on the same page.
I was moved to cloud.hosthatch.com and effectively have less disk space (around 800GB less) now due to a difference in calculation of 10TB.
Compared to
On the other hand, I got a reply in 5 days. Not too bad.
Nah, I quoted directly "I'm just saying that it's somewhat expected to get less hand-holding and less support when things go wrong the lower in price you go, in general across all low end hosts" and responded that instead of less hand-holding and less support, there isn't any hand-holding or any support even when host fuck up and solution depend solely from them.
This storage data corruption is just one of many hosthatch fuck-ups if you read this forum and my response isn't limited to the present one.
From the sounds of it though, people's disk images are corrupt and it's being blamed on a known RAID controller issue. With no real confidence that the disk corruption won't happen again, it seems less about people asking for support and more about being hosted on a server that actually works.
Yeah I absolutely agree with you.
Have you tried using "tune2fs -m" to reduce the amount of reserved space in the filesystem? Personally I'd never set it below 1%, but it's up to you. The default is 5%, which probably accounts for some of the difference, and the rest is probably 1000 vs 1024 in the calculations...
TRIMMED
@DP I didn't say that - so either my account here is hacked or you messed up the quoting. Please confirm or point me to the post where I said this.
Sorry but I probably quoted your quote, LOL.
I'll update that.
EDIT: Fixed and thanks for pointing it out
@dp thanks, I flagged it because I wasn't sure you were on. Please see flag for details if necessary. Thanks for adjusting.