Howdy, Stranger!

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


Do you have a problem with ZxHost? - Page 7
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.

Do you have a problem with ZxHost?

145791034

Comments

  • Only ftp Upload. I guess you can upload via ftp command from cli. Other than that no. You dont get some cutting edge features :P Just plain old File Hosting^^

  • @emptyPD said:
    at leats he can try to restore critical files from clients like me,

    What makes you so special and your data more critical than everyone else's?

    Thanked by 2willie svmo
  •  Hello,
    
    At this point it look's like CEPH are not going to have a bug fix for this issue in a timely manor.
    
    Due to this we are going to start to bring the VM's online with what we can access, this will take some time and will require some manual work on most VM's to bring these back online.
    
    If you see your VM booting please do not send stop  & start commands as this will only cause further delays & issues.
    
    Thanks,
    ZXHost

    Fingers crossing time

  • svmosvmo Member
    edited November 2017

    OK - so it is plan B

  • Good luck to anyone affected!

  • FalzoFalzo Member
    edited November 2017

    I'd like to ask all people trying to help @AshleyUK and of course other clients to get easily through this.

    if you don't need to access or move out your data instantly, please be reasonable and give it some additional time (so do I). please don't start your rsyncs or torrents or whatever right away if not totally needed, to give those with critical stuff a chance to recover what they need better/faster.

    that said, please also think about closing your open tickets regarding this incidents that actually don't really require an answer yet (like asking for what's happening, for an ETA, further updates and so on) just to spare on Ashleys time to go through a big backlog of outdated questions afterwards...

  • Mr_TomMr_Tom Member, Host Rep

    @Falzo said:
    I'd like to ask all people trying to help @AshleyUK and of course other clients to get easily through this.

    if you don't need to access or move out your data instantly, please be reasonable and give it some additional time (so do I). please don't start your rsyncs or torrents or whatever right away if not totally needed, to give those with critical stuff a chance to recover what they need better/faster.

    that said, please also think about closing your open tickets regarding this incidents that actually don't really require an answer yet (like asking for what's happening, for an ETA, further updates and so on) just to spare on Ashleys time to go through a big backlog of outdated questions afterwards...

    Well said!

    Might help prevent it going down with more IO issues after everyone suddenly hits their VM to get data.

  • @Mr_Tom said:

    Might help prevent it going down with more IO issues after everyone suddenly hits their VM to get data.

    exactly. some others already mentioned that they wouldn't neccessarily need their data restored, so same goes for most of mine. and I won't touch anything during the next few hours or days if not needed to give it some room to hopefully stabilize.

  • Mr_TomMr_Tom Member, Host Rep

    @Falzo said:

    exactly. some others already mentioned that they wouldn't neccessarily need their data restored, so same goes for most of mine. and I won't touch anything during the next few hours or days if not needed to give it some room to hopefully stabilize.

    I did reply to one of the support emails. Most of whats on mine is replicated, but there is some stuff on the main disc I could do with. If CEPH had hard disc image files (I've not looked at it) I have said I'd be happy to accept that if it makes it easier.

    Just waiting and keeping an eye on it coming back online at the minute :)

    Thanked by 1Falzo
  • FalzoFalzo Member
    edited November 2017

    another update on what might be happening at this moment: I can see two out of three affected services being able to display the proxmox data in the client area. they are still stopped but it means progress because the hostnodes those are on are up already.
    remember Ashley asked to NOT try and start or stop them yourself ;-)

    probably he goes through all VMs individually, starting them each by hand while monitoring the console during the bootup...

  • Falzo said: probably he goes through all VMs individually, starting them each by hand while monitoring the console during the bootup...

    Here comes the "I don't pay this dude enough to do this" moment.

    Thanked by 2Falzo svmo
  • Mr_TomMr_Tom Member, Host Rep

    @Falzo said:
    remember Ashley asked to NOT try and start or stop them yourself ;-)

    Yeah. I'm in no rush so just waiting now. I do admit I was a bit "aaah" when it went down and updates were a bit sparse but now more information is out I fully understand.

    Just had a look and I'm still seeing no route to host under server information.

  • svmosvmo Member
    edited November 2017

    Looks like Ashley is still trying to raise some of the developers - but the time zone difference might be 8 hours .... so patience is advised !

    I can see my machine in the client area - and see under task history that it had an error starting. - The node probably tried to austostart the VM once it came up.

    So hands off until Ashley brings it up !

    @caracal said:

    Falzo said: probably he goes through all VMs individually, starting them each by hand while monitoring the console during the bootup...

    Here comes the "I don't pay this dude enough to do this" moment.

    You are so right ! - and I'm definitely not going to reward him by taking my business elsewhere

    • Any refund can stay in my account and go towards a replacement service, should it become available.
    • Otherwise I will use it on another of the offers he comes up with from time to time.
  • Mr_TomMr_Tom Member, Host Rep

    @svmo said:

    @caracal said:

    Falzo said: probably he goes through all VMs individually, starting them each by hand while monitoring the console during the bootup...

    Here comes the "I don't pay this dude enough to do this" moment.

    You are so right ! - and I'm definitely not going to reward him by taking my business elsewhere

    • Any refund can stay in my account and go towards a replacement service, should it become available.
    • Otherwise I will use it on another of the offers he comes up with from time to time.

    True!

    I had been considering one of zx's other VPS services anyway.

  • NeoonNeoon Community Contributor, Veteran

    Gluster works fine with big files, if you run a LXC Container on it, you can wait ages.

    Even with SSD/NVMe and 3ms latency between the servers.

    But CEPH, it shat itself after 1 week, I never lost something at Gluster or it crashed.

    I do recommend TINC + GlusterFS works fine.

    Thanked by 2svmo nessie
  • has anyone had their vps come back online?

  • emptyPDemptyPD Member
    edited November 2017

    @elofty said:
    has anyone had their vps come back online?

    one of my two is in "Running" state but is like hang. i dont have access, the other is still "HTTP/1.1 595 No route to host"

    im not touching anything as the email instructions, hope he can solve.

  • two on 'stopped' state here and one 'no route to host' - so nothing changed so far.

  • Mr_TomMr_Tom Member, Host Rep

    Mines still on no route to host.

  • dfroedfroe Member, Host Rep

    I am feelin' with you. "HTTP/1.1 595 No route to host" is all I have... Not going to touch anything - besides my F5 key every other hour. :)

  • same here - one in 'stopped' state

  • Please start MyZXHostVPSStatus group so people can twat about it every 5 seconds.

  • @WSS said:
    Please start MyZXHostVPSStatus group so people can twat about it every 5 seconds.

    I think that thread here is doing just fine :-P

  • MasonRMasonR Community Contributor

    @Falzo said:

    @WSS said:
    Please start MyZXHostVPSStatus group so people can twat about it every 5 seconds.

    I think that thread here is doing just fine :-P

    Ashley should point his status page to this thread :P

    Thanked by 2Falzo imok
  • lurchlurch Member
    edited November 2017

    Just checked mine as yesterday showed no route to host. CP says Uptime 09:31:36 ? but doesn't respond to ping. I'm not touching it especially as Ashley made a point of saying not to.

  • @lurch said:
    Just checked mine as yesterday showed no route to host. CP says Uptime 09:31:36 ? but doesn't respond to ping. I'm not touching it especially as Ashley made a point of saying not to.

    same as one of mine 09:47:49 but no ping.

  • letboxletbox Member, Patron Provider
    edited November 2017

    Not tried Ceph before but wonder what replications setting he is using?!

  • had a short glimpse into the status and as far as I can tell not much changed.

    the low RAM usage suggest no VMs running, CPU usage might be a hint to sync/rebuilding activity, still one node down. it's been looking the same since yesterday...

    I think Ashley still tried to recover something... will see if he gives another update via mail this morning.

    most likely it is what is it and he already stated on the ceph mailing list yesterday:

    However right now the cluster it self is pretty much toast due to the amount of OSD's now with this assert.

    I'd say time to ask people if they really want him to continue to recover data or if it is fine to make a cut and throw it all away...

  • Falzo said: I'd say time to ask people if they really want him to continue to recover data or if it is fine to make a cut and throw it all away...

    Throw it all away? Wasn't only a small amount of data affected, so that >95% of the machines could run? As mentioned before at least for me a "normal" VM is affected and not just backup storage. I would really really like to get this VM back :/

  • chrisp said: I would really really like to get this VM back :/

    while I understand your situation (hence I wrote he should ask and try to filter out the once willing to drop it), how long do you expect him to try (and others to wait) before a final cut should be made? no offense, serious question.

    afaik the problem is that the cluster won't stay online properly even though only a single disk had issues in the first place. because after replacing that, others dropped out or at least lost peering. I am not into ceph but I think it's very hard to work around it or even access the data that's still intact.

    or maybe he is just about to backup all data/disk-images out to somewhere else, after that setting up the cluster from scratch and restoring them. the switch still seems congested, which might support such a scenario and that probably is the best way to do it, if he gets the cluster online at all.

    but esp. for that case I stand by my statement that it would have helped and probably sped up things a lot, if he'd asked clients first what services would be needed to try and recover and which could have just been ignored/removed. to transfer some TB or not can make a little big difference ;-)

Sign In or Register to comment.