Maintenance down time, Never touch a running system? Options

[schnoog]

Hi folks,

6 and a half years ago I was asked to become admin of the Splatterladder.
During the last 6 years we had a total downtime of 51 hours, caused by several defects like broken hard disks (in total 9 were destroyed yet), a broken power supply or a runaway update.
A lot of work was necessary to patch an outdated operating system to keep it in a working and safe state.
But now the time has come to upgrade the operating system to ensure safety and continuity without compiling everything from source.

The SL codebase was written a long time ago, and I'm not able to guess how long it will take to bring all services back to life.
I'll do my very best to minimize the interruption(s).


Greetz
Schnoog

[schnoog]

Thank you all for the flowers :)

But I need to compliment our hoster. Beside the fact that round about 4k Euro flow to them during this time (approx. 1k through donations), the support was always the best.
Never had to wait to exchange any hardware for longer than 4 hours.

If you like some statistics:
SL has moved 3times during the last 6 years. (Hetzner -> Hetzner -> Server4You)
During those 6 years I was connected for 613 hours via SSH to the server, today the 2300th database backup was created (today: 28GB database backup).
Our backup-drive now has a capacity of 3TB which is filled up to 90%

[Wolf Enstein]

.....today the 2300th database backup was created (today: 28GB database backup).
Our backup-drive now has a capacity of 3TB which is filled up to 90%

Good grief .