Re: [Sks-devel] problem with Debian squeeze

[Kiss Gabor (Bitman)]

> > After Debian lenny->squeeze upgrade I got this:
> > 
> > /etc/cron.daily/sks:
> > db4.7_archive: Program version 4.7 doesn't match environment version 4.6
> > db4.7_archive: DB_ENV->open: DB_VERSION_MISMATCH: Database environment 
> > version mismatch
> > run-parts: /etc/cron.daily/sks exited with return code 1
> > 
> > What to do now?
> > I've little experience with Berkeley DB.
> > Any hint will be appreciated.
> > 
> > SKS was upgraded from 1.1.0-4 to 1.1.1+dpkgv3-6.
> 
> Based on the docs at the BerkelyDB (now Oracle) site, this is the process I've
> used each time I have upgraded bdb versions. The only change from mine is to
> Debianize the dir and command names.
> 
> cd /var/lib/sks
> for db in DB PTree
> do
>   db4.6_checkpoint -1h $db  # make sure everything is committed to the db
>   db4.6_recover     -h $db  # remove the old version environment
>   db4.7_checkpoint -1h $db  # This converts the log to the new version
>   db4.7_archive    -dh $db  # remove old log files
> done
> 
> The __db.00[1-5] environment files will be recreated when sks restarts.

Thank you, John! :)

Unfortunately after repairing database I got new error messages:

address@hidden:/etc/sks$ /usr/sbin/sks db
Fatal error: exception Keydb.Unsafe.No_db
address@hidden:/etc/sks$ /usr/sbin/sks recon
Fatal error: exception Bdb.DBError("unable to join the environment")
address@hidden:/etc/sks$ 

According to strace this happens right after opening /var/lib/sks/DB/__db.001:

[...]
open("/var/lib/sks/DB/__db.rep.init", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No 
such file or directory)
open("/var/lib/sks/DB/__db.001", O_RDWR|O_LARGEFILE) = 7
fcntl64(7, F_GETFD)                     = 0
fcntl64(7, F_SETFD, FD_CLOEXEC)         = 0
fstat64(7, {st_mode=S_IFREG|0600, st_size=24576, ...}) = 0
close(7)                                = 0
open("/var/lib/sks/DB/__db.001", O_RDWR|O_LARGEFILE) = 7
fcntl64(7, F_GETFD)                     = 0
fcntl64(7, F_SETFD, FD_CLOEXEC)         = 0
mmap2(NULL, 24576, PROT_READ|PROT_WRITE, MAP_SHARED, 7, 0) = 0xb7712000
close(7)                                = 0
munmap(0xb7712000, 24576)               = 0
time(NULL)                              = 1291813181
stat64("/etc/localtime", {st_mode=S_IFREG|0644, st_size=866, ...}) = 0
write(4, "2010-12-08 13:59:41 Shutting dow"..., 43) = 43
write(2, "Fatal error: exception Keydb.Uns"..., 42Fatal error: exception 
Keydb.Unsafe.No_db
) = 42
exit_group(2)                           = ?

(Note that __db.001 was not read but opened only.)

What should I do? :-/

Regards

Gabor
-- 
A tavalyi bajnok reszvetele nelkul indul a Masodik Nemzetkozi Kamikazeverseny