YaCy-Bugtracker - YaCy
View Issue Details
0000770YaCy[All Projects] Generalpublic2017-10-06 05:572019-07-28 08:38
smokingwheels 
 
normalcrashunable to reproduce
newopen 
none 
Raspbian jessie
YaCy 1.9 
 
0000770: Excess tlog files in index folder.
A Peer had crashed some time ago.
I was sshed in and was in the process of erase when I noticed 1000's of tlog files being erase, I closed bitvise and checked it out there was over 20 000 of them.
I took a copy and pasted the index folder into the DATA folder of a new clone on another PC.
It started it and it worked, I set the config up again.
 
I scheduled all the basic configuration to happen every 10 min also optimize solr segments.
I also upgraded the Pi's firmware sudo rpi-update There are issues with it but the main thing is that my HDD io transfer has improved.

Rename dead peer and clone source again.
Copy index folder from old peer to inside clone's DATA folder without starting it.
Start and reconfigure settings.

No tags attached.
Issue History
2017-10-06 05:57smokingwheelsNew Issue
2017-10-06 13:28smokingwheelsNote Added: 0001483

Notes
(0001483)
smokingwheels   
2017-10-06 13:28   
If You have an older peer to rename the freeworld SEGMENTS solr_5_5 to solr_6_6.

I found The first time starting it the default memory settings ended in GC overhead limit exceeded so I killed the task and edited yacy.conf to increase Xmx and Xms and tried again and my old peer 4040 started working.