Server up again, let's see for how long it will last this time...
EDIT: Gah, it went down again... -.-" However, I found out that this is a known problem in Abe related to concurrent blockchain processing when new block is received from the network (race condition or something). I've implemented a quick fix - using a separate process for writing new blocks into DB once in a while (so webpage might be delayed by up to 60 seconds, not a big deal). Hopefully this avoids another DB corruption until proper fix is implemented in Abe.
Ahh, my young ABE Padawan.
The first step in your quest for a smooth running explorer is to switch to the FastCGI server and stop using the default built in python server.
You have the right idea about separate processes for update verses serve. Keep two copies of abe.py with different names like abe.py and uabe.py.
Any requests from the web server will use one abe.py with a conf file that uses a empty bracket for datadir (datadir = [] ). This will stop web requests from tripping an update.
To update the database use the blocknotify option in the coin daemon conf file and point it to a script that will run the second abe.py (uabe.py) with a full conf file to trigger a database update. The reason for the two named abe.py has to do with FastCGI sticking its big nose into your database update process if it has the same name as the process it trips.
If you need any more help, just let me know.
! thank you very much and very sorry for your time!!
