Sonntag, 31. August 2014
Samstag, 30. August 2014
Freitag, 29. August 2014
Donnerstag, 28. August 2014
Mittwoch, 27. August 2014
Sonntag, 24. August 2014
Samstag, 23. August 2014
Freitag, 22. August 2014
Donnerstag, 21. August 2014
Mittwoch, 20. August 2014
Dienstag, 19. August 2014
Montag, 18. August 2014
Sonntag, 17. August 2014
Samstag, 16. August 2014
Freitag, 15. August 2014
Mittwoch, 13. August 2014
Dienstag, 12. August 2014
Montag, 11. August 2014
Samstag, 9. August 2014
Freitag, 8. August 2014
Mittwoch, 6. August 2014
Dienstag, 5. August 2014
Machine Down (did the Samsung 840 Pro SSD crash the Ubunt 14.04 server?)
Since the hardware was old (Core2 Duo) I upgraded my server a couple of month ago to a Core i7-4771 on an Asus 90MB0DZ0-M0EAY5 mainboard with 32GB Kingston KVR16E11K4/32 memory. And, since I had good experiences with an Samsung 830 Pro SSD, I added a Samsung 840 Pro SSD.
I was very happy with the setup - everything worked well.
Until Monday: When I tried to access the server on Monday morning it did not respond and when I checked, I found it had restarted and was hanging in the kernel selection. So I pressed enter, to initiate the boot sequence, but soon after it had started to boot, the screen when blank and power was gone.
I was not able to start it again with the hardware reset button or after powering it down with the power button.
After a couple of tries I managed to start it again (which involved switching off the power supply for some time and disconnecting the SSD).
Whenever I got it to boot, it shutdown very soon after the boot sequence had started.
So I feared the worst, weight my options and tried to isolate the problem.
Equipped with a USB stick containing an Ubuntu server installation image and the old hard drive, I managed to get the server running again with the hard drive and no more mysterious shut downs.
I even got it to work with the SSD again, but it turned out, only when the SSD and the hard drive were both installed. Accessing the SSD seemed no problem in such a constellation and I was abel to backup the latest data.
I then let it run for a couple of hours with SSD and hard drive (the hard drive not used, it just seemed to make things stable) and thought that this combination was OK.
Until this morning I noticed that the server was down again.
So I removed the SSD and just let the server run with the hard disk while I was out for work today.
Since it was still OK when I got home, I copied the root partition from the SSD to the hard drive and (after reinstalling GRUB) managed to boot from the hard drive with a working, fully installed system.
I then removed the SSD and the system is running now.
If it makes it through the night (and maybe a couple of days) I can rule out other defects and replace the SSD. But for now I'm just hoping that it is stable until tomorrow...
I was very happy with the setup - everything worked well.
Until Monday: When I tried to access the server on Monday morning it did not respond and when I checked, I found it had restarted and was hanging in the kernel selection. So I pressed enter, to initiate the boot sequence, but soon after it had started to boot, the screen when blank and power was gone.
I was not able to start it again with the hardware reset button or after powering it down with the power button.
After a couple of tries I managed to start it again (which involved switching off the power supply for some time and disconnecting the SSD).
Whenever I got it to boot, it shutdown very soon after the boot sequence had started.
So I feared the worst, weight my options and tried to isolate the problem.
Equipped with a USB stick containing an Ubuntu server installation image and the old hard drive, I managed to get the server running again with the hard drive and no more mysterious shut downs.
I even got it to work with the SSD again, but it turned out, only when the SSD and the hard drive were both installed. Accessing the SSD seemed no problem in such a constellation and I was abel to backup the latest data.
I then let it run for a couple of hours with SSD and hard drive (the hard drive not used, it just seemed to make things stable) and thought that this combination was OK.
Until this morning I noticed that the server was down again.
So I removed the SSD and just let the server run with the hard disk while I was out for work today.
Since it was still OK when I got home, I copied the root partition from the SSD to the hard drive and (after reinstalling GRUB) managed to boot from the hard drive with a working, fully installed system.
I then removed the SSD and the system is running now.
If it makes it through the night (and maybe a couple of days) I can rule out other defects and replace the SSD. But for now I'm just hoping that it is stable until tomorrow...
Samstag, 2. August 2014
Freitag, 1. August 2014
Abonnieren
Posts (Atom)