With new server OS the Boinc software did not compile good. It need more time to make it running. However due to holiday period, this cannot be complete until September.
Copy of data from old server to new is fnished. Now it is being configuring BOINC related services.
The process of moving all the BOINC archive from old server to new server (locate remotely) is still proceding.
Big news: thanks to Reese the project can continue with a new server with RAID10 disks, so disk failure will not stop the server.
As the server is locate remotely, it will need many times to set up it and transfer data from old to new server, as both old and new server must be prepared for the operation.
So attend a BOINC project update message when all will be ready.
Thanks.
All disks where tested offline and all where in good state, but in server they shows errors on boot. While I was suspecting an hardware controller problem on the server, the root disk did not respond anymore, even in offline pc!
So it is probably that the M2 SSD disk starts to have some performance problems that slow down the server activity and stops BOINC normal activities, but those problems did not appears when mounted into the PC (it runs fast).
But now it died in a moment, and as this is a new disk with less that 10 months of activities, this implied that disk quality is a big problem!
For this it is very probably that the ODLK1 project will shutdown: changing a disk every 9 months to prevent future problems is very destructive.
So at least, it will be restored the board with many results provided during the years, and the latest result will be processed to be release to the public.
Very sad day for the project.
However many thanks to all that has process the WU during all those years for providing many good result into the matemathical problem of latin square.
The server had a process that blocked normal execution of it for some times. Now I'm investigated the issue (for software or hardware problem)
UPDATE: after some investigation, I need to check the disks offline, so the server will be power off during that operation.
Server is running in "emergency mode" and onto the weekend it will be power off for fixing it's automatic boot up failure
There was a lightning bolt in the night that damage many electrical devices in the area. Internet connection were restored in the mornung, now server is being checked if it has got damage or not.
Both root and data disks on the server are good (tested mounted in another machine), but actually it did not bootup.
The problems will take many days to investigate, so use a backup BOINC project. Thanks.
The server is having some sporadic problems on webservice. Be sure to have some backup project to use, as the problem will be investigated during the next weekend.
During the night the power missed for more that 6 hours and power supply did not last so long.
When the server start for power returns, not all disk were mounted, so only after a proper reboot it start to operate again at 18:35.
Server operation is now 100% up.
Forum is active again. Many manual actions where done to fix the filesystem with DB in order to boinc job to run again.
Actually the restore of client downloading is not active
Unfortunately the new SO has setup/libraries imcompatibility with previous one, so the old used setting is not still valid.
This require to write from scratch all the Boinc related stuff, so I will leave the server sconnected from internet when working on this.
The DB actually is the one of 20 Aug at 8.00, the disk problems so come after that date.
SO was installed during the weeek. Now the Boinc DB is popolated with the DB bump, and then all services will be recreated.
Disk is arrived today and it is installed. Now the installation of new SO can begin
During the vacation period the server had an hardware failure on the M2 SSD disk (as I was not connected to server during this time, a possible pre-failure signal in log was not seen). It is the one with the fast access for the SO partition, not the BOINC data (that is in an external big SSD disk).
Maybe failure happened during 20 to 23 August, and probably the system hang up as when physical connected to it, it as no one response.
Now a new disk is being orderer and then the SO will be installed (it will be used the more recent LTE version). I hope to have some news soon.
Thanks