Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Hetzner Cloud very unstable! - Page 2
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.

Hetzner Cloud very unstable!

24

Comments

  • JordJord Moderator, Host Rep

    I haven't had an issue with their cloud product yet, maybe I'm on one of the lucky nodes.

  • They seem to have gone through a power failure that has affected me.

  • ArchieArchie Member

    My cloud server is down for hours for no reason and no reply from support ticket

  • MikePTMikePT Moderator, Patron Provider, Veteran

    @Archie said:
    My cloud server is down for hours for no reason and no reply from support ticket

    Happened to me a week ago, just had to shutdown/boot from their panel, have you tried it?

  • I checked mine:

    nbg1-dc3

    up 11 days, 4:35

    Working fine but I can't see anything in my Hetzner panel. No graphs at all!

  • erkinerkin Member

    @neik said:
    Hi guys,

    currently I am on Scaleway but after a couple of problems with it I am getting sick of the unexpected downtime. Besides that, I need more than 2GB RAM and this is where Hetzner/Netcup (price-wise) come into play.

    Their 4GB RAM VPS are almost the same price for me (german resident) so that I am wondering which one I should give a try first?

    Has anyone had experiences with both and can do a little comparison (stability, downtime and in case you need it: support)?

    Thanks!

    Never tried netcup and no problems with hetzner until now.

  • ArchieArchie Member

    @MikePT said:

    @Archie said:
    My cloud server is down for hours for no reason and no reply from support ticket

    Happened to me a week ago, just had to shutdown/boot from their panel, have you tried it?

    I find those on the status page which might be the cause but define how much urgent it is, doing reboot without notification is a very bad experience for new customers like me, imagine the uptime can not up to 7 days, what kind of service you would put on it

    https://hetzner-status.de/#8920
    https://hetzner-status.de/#8919
    https://hetzner-status.de/#8917

  • MikePTMikePT Moderator, Patron Provider, Veteran

    @Archie said:

    @MikePT said:

    @Archie said:
    My cloud server is down for hours for no reason and no reply from support ticket

    Happened to me a week ago, just had to shutdown/boot from their panel, have you tried it?

    I find those on the status page which might be the cause but define how much urgent it is, doing reboot without notification is a very bad experience for new customers like me, imagine the uptime can not up to 7 days, what kind of service you would put on it

    https://hetzner-status.de/#8920
    https://hetzner-status.de/#8919
    https://hetzner-status.de/#8917

    Well its no good. But hw issues do happen.

  • @MikePT said:

    @Archie said:

    @MikePT said:

    @Archie said:
    My cloud server is down for hours for no reason and no reply from support ticket

    Happened to me a week ago, just had to shutdown/boot from their panel, have you tried it?

    I find those on the status page which might be the cause but define how much urgent it is, doing reboot without notification is a very bad experience for new customers like me, imagine the uptime can not up to 7 days, what kind of service you would put on it

    https://hetzner-status.de/#8920
    https://hetzner-status.de/#8919
    https://hetzner-status.de/#8917

    Well its no good. But hw issues do happen.

    happen to often there :) as i see

    I remember, netcup has hw issue too, around once per year.

  • Facing slow speed issue with new deployments. Old CX41 seems to be working fine, but I had a need for a new project, for which a CX21 and a CX41 was deployed, but it seems to be crawling right out of the box.

    My old VPS can download at 150MBps and the new ones are at around 38MBps

    Disk looks like very slow too.

    See attached.

    image

  • ZerpyZerpy Member

    @realbusiness said:
    Facing slow speed issue with new deployments. Old CX41 seems to be working fine, but I had a need for a new project, for which a CX21 and a CX41 was deployed, but it seems to be crawling right out of the box.

    My old VPS can download at 150MBps and the new ones are at around 38MBps

    So 376.8megabit/s is simply too little for you? Try using another vultr location - or some other speed test option, there's plenty of them around the web.

    Thanked by 1vimalware
  • @Zerpy said:

    @realbusiness said:
    Facing slow speed issue with new deployments. Old CX41 seems to be working fine, but I had a need for a new project, for which a CX21 and a CX41 was deployed, but it seems to be crawling right out of the box.

    My old VPS can download at 150MBps and the new ones are at around 38MBps

    So 376.8megabit/s is simply too little for you? Try using another vultr location - or some other speed test option, there's plenty of them around the web.

    I am simply comparing 3 months old vs today's deployment, i.e the ones I have deployed today, seem very slow. Slow disk and slow downloads. I have tried a few locations, old vps is running many times faster on all locations tried, not just from Vultr, but from others too.

    PS: I use a multiple of 8 to convert byte to bit, is that wrong?

  • mkshmksh Member

    @realbusiness said:
    PS: I use a multiple of 8 to convert byte to bit, is that wrong?

    Tbh i don't i don't know how you arrive at those numbers at all. Your screenshoot shows 47MB/s which is about 500mbit/s. Hint: Use wget -O /dev/null http://... to download to avoid the disk as a bottleneck.

  • mkshmksh Member
    edited June 2018

    Yeah but given your screenshot shows 47.1MB/s how do you arrive 38mbit/s or is it a typo and you just ment 48MB/s rounded up?

  • I used the average speed, which is 38.3

  • williewillie Member

    realbusiness said:

    I used the average speed, which is 38.3

    That's over 300 mbit which is fine. 1200+ is simply insane.

  • ZerpyZerpy Member

    @realbusiness said:
    I used the average speed, which is 38.3

    Which is 306.4 mbps - plenty for the price you pay.

  • mkshmksh Member

    @realbusiness said:
    I used the average speed, which is 38.3

    I see. My bad then.

  • realbusinessrealbusiness Member
    edited June 2018

    New server deployments are averaging 3mins each!

    I think its time for a new provider!

  • ZerpyZerpy Member

    @realbusiness said:
    New server deployments are averaging 3mins each!

    I think its time for a new provider!

    hmm, not sure where you're deploying, but I just spawned 3 instances, and they all came up within like 25 seconds from clicking "deploy" to the machine actually pinging.

  • CX41 - 3 mins to deploy - 5+ mins to rebuild (I am testing RunCloud / CMM)

  • williewillie Member

    If you mean restoring from a snapshot, then yes, that does take longer than launching a bare server.

    Thanked by 2Aidan vimalware
  • No snapshot, simple Rebuild.

  • avelineaveline Member, Patron Provider

    I also had an instance in Hetzner Cloud.

    It reboots without notice twice last week.

  • williewillie Member
    edited June 2018

    realbusiness said:

    No snapshot, simple Rebuild.

    You mean reinstall all packages? Yes of course that takes a while.

    aveline said:

    It reboots without notice twice last week.

    If that was the power fail incident, yeah there was another thread about it. If it was something else, then hmm that's more of a concern.

    I do think the early performance of this product was just insane, i.e. they deployed a ton of hardware on day 1, and it took a while for the user population to ramp up. So the amount of hardware per user was ridiculously high for a product at this price level. I remember trying it out and being amazed. It's unsurprising if that turned out to be too good to last. So users who set their expectations right after the product launch probably were being unrealistic.

    I haven't tried it again recently for anything substantial but might do so soon. I do have a CX11 instance running right now that seems to work fine, but it's just idling or running a fairly light workload. I'll have something heavier for it soon. I still don't think there's anything else comparably priced to the bigger instances.

  • ZerpyZerpy Member
    edited June 2018

    I happily run a MySQL slave of a 25 gigabyte database on it (CX21) - it does perform amazing - to be honest, it's 5 euros ... I think people expect too much sometimes :|

    Thanked by 1iKeyZ
  • AkaikeAkaike Member
    edited June 2018

    I have the same problem. It happened already twice, my servers are in Falkenstein and I'm not sure if there was an electricity problem. But a few weeks ago it turned off itself without any warning. And today again and it didn't turn on by itself, I had to do it after I realized that the server is offline for no reason.

    It's cheap and very good performance wise but I'm not sure about the stability yet.

  • openosopenos Member

    nbg1-dc3

    DKoTi.md.png

    Thanked by 1nfn
  • Deployed some new CX21 today and it's flying!

Sign In or Register to comment.