Page 1 of 1
0.5.0 crc error worse than 0.4.12
Posted: December 20th, 2009, 5:12 am
by priesh
Hey - Got an nzb to test the new beta - on 0.5.0beta4 it errored, and 12 out ofthe 23 video files had extracted successfully in the resulting failed folder. I tried the same nzb on the older 0.4.12 (uninstalled the new version and installed the old one) and it also got a crc error, but 22 out of the 23 files extracted successfully.
Interestingly the 12 that extracted were episodes 1-12 for 0.5.0beta4, whereas the one file that failed on 0.4.12 was the 20th episode. Using quickpar on the 20th episode also failed so the old version worked properly.
Re: 0.5.0 crc error worse than 0.4.12
Posted: December 20th, 2009, 5:20 am
by shypike
Can you email the NZB file to bugs at sabnzbd.org ?
Re: 0.5.0 crc error worse than 0.4.12
Posted: December 20th, 2009, 5:32 am
by priesh
sure - done.
Re: 0.5.0 crc error worse than 0.4.12
Posted: December 20th, 2009, 8:02 am
by shypike
The problem is that one set doesn't verify properly.
When you have the option "Post-process only verified jobs" on, no unpacking will be done.
When the option is off, unpacking will start, but will stop after the first failing set.
The ordering of the sets is different for 0.4.12 and 0.5.0, hence the different results.
Re: 0.5.0 crc error worse than 0.4.12
Posted: December 20th, 2009, 10:25 am
by priesh
Ok thanks for the info.
Can't it be set so it unpacks all verified sets and just leaves the unverified one(s) out? So 0.5.0 would unpack the 22 good ones and leave the bad one?
Also i dont know if this is related but if i d/l a massive nzb (80gb) which contains a few videos, then it will always fail with a crc error (0.5.0b4) but if i get nzbs for each video individually then it works fine?
Thanks - awesome prog btw!
Re: 0.5.0 crc error worse than 0.4.12
Posted: December 20th, 2009, 4:57 pm
by shypike
Your guess is correct.
We are working on giving the sets within an NZB separate treatment,
but this will not be in 0.5.0 (or we will never finish it).
Re: 0.5.0 crc error worse than 0.4.12
Posted: December 21st, 2009, 3:57 am
by priesh
Right that makes sense - so would it be better to revert 0.5.0 to use the same set ordering as 0.4.12, or will 0.5.0 on average do as well as 0.4.12?