Topic: SME 10 asks "Restore from backup" after normal power off/on

[Skalviai]

Hi.

Help.
After shutting down the SME server, the server starts up and asks "Restore from backup", "Choose an administrator password", ....

What to do?

[TerryF]

Is this the same system that was subjected to power cycling and was experiencing this same situation on power up?

[ReetP]

I think the OP needs to take a few steps back and tell us what happened before.

History, hardware etc....

I presume they also have a good current backup....

[ReetP]

Is this the same system that was subjected to power cycling and was experiencing this same situation on power up?

Ahh this old chestnut?

https://forums.koozali.org/index.php/topic,54633.0.html

[Skalviai]

Is this the same system that was subjected to power cycling and was experiencing this same situation on power up?

No.
This server worked for 5 years with SME9.2 and migrated (new installation) to SME 10.0
The server shuts down on Friday and turns on on Monday.

[Skalviai]

Ahh this old chestnut?

https://forums.koozali.org/index.php/topic,54633.0.html

The situation looks similar, only this time the server was shut down correctly.

[TerryF]

No.
This server worked for 5 years with SME9.2 and migrated (new installation) to SME 10.0

So clean install of sme10 and restored a backup, which backup and restore method was used

The server shuts down on Friday and turns on on Monday.

So SME10 had been running fine up until Friday, all recent updates had been performed as they arose, system had been through recent multi restarts and reconfigs with those updates before it was powered down on Friday and then restarted Monday..you do need to really paint the picture for us to try and get a understanding of what is happening, befiore and after.

[ReetP]

Clear concise history please.

How to report issues effectively:

https://forums.koozali.org/index.php/topic,54724.0.html

[Jean-Philippe Pialasse]

system password routine starts with

Code: [Select]

    return if ($db->get_value('PasswordSet') eq 'yes');
perform restore with

Code: [Select]


    return if (($db->get_prop('bootstrap-console', 'Run') eq 'no') && $self->{bootstrap} ); # regular reboot
    if ($db->get_prop('bootstrap-console', 'Run') eq 'yes') # called from bootstrap console
    {
        return if ($db->get_value('PasswordSet') eq 'yes'); # too late
    }
   

meaning if PasswordSet key is
- not set
- or set to yes
- or /home/e-smith/db/configuration is corrupted
- or /home/e-smith/db/configuration is inaccessible then those  two tasks will be launched on restart. 


situations i can see:
- file corrupted because of power down when it was write
- file corrupted because manually edited
- one changed to yes
- /home/ or /home/e-smith or /home/e-smith/db has been put on a different block device than the system and was not mounted on boot because of an incomplete setting of grub.


also as both are asked i can say that also that probably one could have run this event
signal-event post-install

or

signal-event post-upgrade ; and manually changed the SystemPassword key too.

then let the server running until the power got down and finally cycle through a reboot.  The post-install is designed only to be used by anaconda and will lock you exactly that way on reboot.

pick the one that seems most probable to you. 

[Skalviai]

So SME10 had been running fine up until Friday, all recent updates had been performed as they arose, system had been through recent multi restarts and reconfigs with those updates before it was powered down on Friday and then restarted Monday..you do need to really paint the picture for us to try and get a understanding of what is happening, befiore and after.

The server migrated to version 10 in November, and was shut down every Friday, and restarted on Monday.
"Restore from backup" message when turning on the server this Monday.

[ReetP]

So we have no more information than when you first started this thread, despite asking.

We can't help you if you can't help us.

Please give a more accurate history of updates, restarts etc - what happened and when as best you can remember. And as much information as possible.

What is the hardware layout - from your previous post I think you had a number of different disks? Had one died?

How was it shut down (and why?)?

What updates had been done and when?

Have you tried booting with a USB stick and looking on the partitions for some logs and other clues? Can you see and mount all the partitions at least read only?

Please, start giving us some more information or we can do nothing more to help you.

All you have really said so far is 'it won't boot' and that is it.

JP has outlined the usual reasons for that so unless you have further information there is nothing else we can do apart from suggest you start again and restore from your backup.

You have got a backup haven't you as you haven't answered that question either??

[Skalviai]

The server has a backup, and has already been restored from it. All data is on another drive (RAID1).
No help is needed, just maybe someone has already solved this problem.
Server: CPU i5 4core, 8GB RAM, 120GB SSD (System), 2x3TB SATA disk (RAID1), 750GB SATA disk (backup). All disk work fine.

The server updated only after installation, first  command  after installation - "yum update"

SME server used as a data server and VPN server.

[Jean-Philippe Pialasse]

You are the only one reporting this problem.
so without any way to reproduce this issue we can not fix it.

we asked you detailed information and we did not get them, or only really general information we are just unable to investigate more and as a result you leave yourself all alone with your issue

I even took the time to explain what could be the most probable situation but, as we do not get more information to help you are again alone. 

we are not mind reader, we do not have access to your machine.

help us to help you.