Repairs fail when nzb is reused

[tixetsal]

Hi, folks.

I have 2 different instances of sabnzbd in 2 different cities. In instance 1, sab is fed nzbs by sick/couch/head. After the nzb is processed by sab, it copies (via scp) the nzb from the nzb backup folder to instance 2, so that all media exists in both locations. (Both instances download from the same usenet provider.)

I have observed that many items on instance 2 fail, due to a lack of repair blocks, despite being successfully repaired on instance 1. Does anyone know why this would be the case? Is parity information somehow being stripped before the file is copied?

How could I obtain a "supplemental" nzb to retry without downloading everything over again?

[shypike]

Nothing is removed from the NZB.
Are you using the same Usenet provider in both cities?
Did you set the servers timeouts equal?

[tixetsal]

Thanks for your response, shypike. I am using the same provider in both locations, but I did notice that the timeout was set to 120 seconds in the location that is always successful and 30 seconds in the location that would fail 50-60% of the time.

Good call! Do you think that this could be the source of my problem?

[shypike]

Yes, it could very well be.
We generally recommend 60 seconds.
OTOH if 30 sec is not enough it does make me doubt the Usenet-provider/ISP combination that needs more.