[ovirt-users] Re: Engine dm-1 disk crashed

Hi, We have generate a Script with the following prozedere: https://lists.ovirt.org/pipermail/users/2017-August/083665.html Hosted-storage will be not touched.

Br Marcel Am 5. Mai 2021 10:40:25 MESZ schrieb Yedidyah Bar David <didi(a)redhat.com&gt;: > > On Wed, May 5, 2021 at 10:59 AM marcel d'heureuse <marcel(a)deheureu.se&gt; wrote: >> >> >> Hi, >> >> Sorry, haven't write it. Yes it is a self hosted engine. >> >> Br >> Marcel >> >> Am 5. Mai 2021 07:42:35 MESZ schrieb Yedidyah Bar David <didi(a)redhat.com&gt;: >>> >>> >>> On Wed, May 5, 2021 at 8:32 AM marcel d'heureuse <marcel(a)deheureu.se&gt; wrote: >>>> >>>> >>>> >>>> Hi, >>>> >>>> We shutdown the complete ovirt system, if a poweroutage occurs and we don't want to make an hard power off. >>>> >>>> So we stop all vms, put the storages in maintenance and shut down the engine, disconnect storages and shut down the nodes. > > > Can you please provide more details? > > Do you follow some existing procedure/document? > > Why do you put storage to maintenance? > > Do you also put the hosted_storage domain to maintenance? If so, this > can explain your fs corruption. If you have valid reasons to put all > other storage to maintenance, perhaps filter out this one. > > You might want to also check this, although I do not know of many > people that use it: > > https://github.com/oVirt/ovirt-ansible-collection/tree/master/roles/shutd... > > Best regards, > >>>> >>>> On power up, sometimes it works fine but in some cases the engine need a xfs_repair session to start. >>>> >>>> What we make wrong? Have you an idea? >>> >>> >>> >>> Is this a hosted-engine? Or standalone? >>> >>> Best regards, > > >