Re: warnings about "Failed to insert for source by digest"
On Wed, 13 Oct 2010, Peter Palfrader wrote:
> > W, [2010-10-12 20:12:17#18500] WARN -- : [indexrun #12] Failed to insert for source by digest, but no file with that digest exists: fsldata 4.1.6-2: fsldata_4.1.6.orig.tar.gz (5e7105a8d989f2eea462a385b4bc56a90d8d4485)
> > so I do not dive into possibly wrong conclusions ;)
> I think this means that you have a .dsc file in your tree that
> references a fsldata_4.1.6.orig.tar.gz with a sha1sum of
> 5e7105a8d989f2eea462a385b4bc56a90d8d4485.
> That file is not to be found in your tree.
> Can you confirm there is no fsldata_4.1.6.orig.tar.gz?
hm... file is there with correct shasum but yesterday was a date of
transition. Michael has removed all fsldata packages from one
distribution and placed it into data distribution later on.
Theoretically this actual could have overlapped with reprepro
manipulations.
I guess I am only yet to setup correct triggering of snapshotting and
possibly locking between reprepro and snapshot since we do not have
time-scheduled "dinstall" runs -- update of the repository is triggered
manually usually right upon upload.
Thus, so far I just cronned full one (inplace import-run neurodebian,
which does simply snapshot import, dumpall and index without any remote
syncing or FUSE) twice a day.
btw, description within mirror/farm-journal-expire says something about
outputing to stdout, but as far as I can see it just operates on DB
without any output actually (or am I wrong? I haven't ran it)
I am yet to comprehend the lean farming mechanism ;)
--
.-.
=------------------------------ /v\ ----------------------------=
Keep in touch // \\ (yoh@|www.)onerussian.com
Yaroslav Halchenko /( )\ ICQ#: 60653192
Linux User ^^-^^ [175555]
Reply to: