[Sks-devel] Re: sks cleandb freezing in 1.0.5 CVS

[Yaron M. Minsky]

This is the bug associated with re-introducing locking that I mentioned
earlier.  I got rid of this problem in my copy, but didn't commit it to
CVS before.   I have committedit now, so if you get the CVS version now
and recompile, you should be good to go.

y

On Thu, 2003-11-20 at 22:31, address@hidden wrote:
> Yaron,
> 
>    after my HDD pooped out on me yesterday I've been rebuilding my 
> database (not due to corruption this time, just  a HDD that decided to 
> give up the ghost, although I don't know if I suspect it of the corruption 
> issues). I wanted to be able to search on SubKey IDs so I grabbed the 
> 1.0.5 from CVS (old 1.0.5 souce was on the dead drive). It took a long 
> time to process fastbuild, which is to be expected, but cleandb is taking 
> HOURS also. I suspect it is stuck waiting for something because ps shows 
> it using 0% of the cpu. I had debug level set to 6, and the output it gave 
> me before seeming to freeze was:
> 
> === Cleaning key database... ===
> 2003-11-20 18:29:54 Opening KeyDB database
> 2003-11-20 18:29:54 Keydb opened
> 2003-11-20 18:29:54 Database already deduped
> 2003-11-20 18:29:54 Merging keys in database
> 2003-11-20 18:29:54 Starting key merge
> 2003-11-20 18:29:54 Multiple keys found with same ID.  merge_from_hashes 
> called
> 2003-11-20 18:29:54 Hash: 4A890E696081B0D3B747143BFB881E79
> 2003-11-20 18:29:54 Hash: 600066ED625D84CE88211E0252A675D7
> 2003-11-20 18:29:55 10 thousand steps processed
> 2003-11-20 18:29:55 Multiple keys found with same ID.  merge_from_hashes 
> called
> 2003-11-20 18:29:55 Hash: 3A2D825000E94AD1FF9CC832CF44CDC5
> 2003-11-20 18:29:55 Hash: E2C79FED8E3889697593DAA8CBCB8A8E
> 2003-11-20 18:29:55 20 thousand steps processed
> 2003-11-20 18:29:55 30 thousand steps processed
> 2003-11-20 18:29:55 40 thousand steps processed
> 2003-11-20 18:29:55 50 thousand steps processed
> 2003-11-20 18:29:56 60 thousand steps processed
> 2003-11-20 18:29:56 Multiple keys found with same ID.  merge_from_hashes 
> called
> 2003-11-20 18:29:56 Hash: 157FCAA6668C0B2779E3A4FA3D9B1BFE
> 2003-11-20 18:29:56 Hash: ABE38837C4EEA542399DE3228F3BB0BF
> 2003-11-20 18:29:56 Multiple keys found with same ID.  merge_from_hashes 
> called
> 2003-11-20 18:29:56 Hash: 110149D0C7EBC3EAD58D5896AB7E09E2
> 2003-11-20 18:29:56 Hash: E828DCB0BD9A2141B9B8281878716B57
> 2003-11-20 18:29:56 70 thousand steps processed
> 2003-11-20 18:29:56 80 thousand steps processed
> 2003-11-20 18:29:56 90 thousand steps processed
> 2003-11-20 18:29:57 100 thousand steps processed
> 2003-11-20 18:29:57 Multiple keys found with same ID.  merge_from_hashes 
> called
> 2003-11-20 18:29:57 Hash: 5149F363CA78C1AE6A34D5A7AF90B169
> 2003-11-20 18:29:57 Hash: 7E085AC797691E3C47835DD7BC5CBBCD
> 2003-11-20 18:29:57 1 replacements found
> 2003-11-20 18:29:57 replacing 2 keys with single merged key
> 2003-11-20 18:29:57 removing: 7E085AC797691E3C47835DD7BC5CBBCD
> 2003-11-20 18:29:57 removing: 5149F363CA78C1AE6A34D5A7AF90B169
> 2003-11-20 18:29:57 adding: 5149F363CA78C1AE6A34D5A7AF90B169
> 
> 
> I'm going to abort it and rerun with a bytecode version of sks, so I can 
> get any other output. Any ideas on this one?
> 
> --- Dan
-- 
|--------/            Yaron M. Minsky              \--------|
|--------\ http://www.cs.cornell.edu/home/yminsky/ /--------|

Open PGP --- KeyID B1FFD916
Fingerprint: 5BF6 83E1 0CE3 1043 95D8 F8D5 9F12 B3A9 B1FF D916