Re: Severe bug in either afio or gpg, but I don't know which
Rob Weir <rweir@softhome.net> writes:
> I have no idea how to fix this, but I've (more than a few times) tarred
> up my MailDir hierarchy, then compressed+encrypted it with GPG (both
> using asymmetric and symmetric ciphers) quite successfully. Any
> particular reason you need to use afio rather than tar?
I used to use tar, but afio handles corrupt data more gracefully, is more
efficient, is able to gpg each file separately, allows plaintext control
files in the same archive as the gpged files, and is designed from the start
around taking a list of pathnames to archive. There is no widespread ecc
program to avert media errors, so gpging each file separately is safer.
cpio is compatible and widely enough available. Also, the args make more
sense to me. My experience; YMMV.
But if it's afio's fault, then I might very well switch to cpio.
I think it's a combination of afio and gpg, though. It happens quite a bit
more on certain files, including /var/lib/dpkg/available. However, it is
intermittent. It never happens when afio does not call gpg.
Script started on Sat Dec 7 22:00:27 2002
$ find available available-old |
afio -oz -Z -U \
-P gpg \
-Q --batch \
-Q --no-options \
-Q --homedir -Q $homedir \
-Q --always-trust \
-Q --no-auto-check-trustdb \
-Q --recipient -Q "${recip:?}" \
-Q --debug-all -Q -vv \
-Q -e \
-3 0 \
"$ofile"
I have the gpg debug output, but it is 1MB and I thought I'd spare people's
mailboxes.
gpg: DBG: fd_cache_open (/root/tmp--xyzzy-nichi/a-tn-qhome-qout--2002-12-07
-Sat-22-00-46-262072/pubring.gpg) not cached
gpg: DBG: iobuf-1.0: open `/root/tmp--xyzzy-nichi/a-tn-qhome-qout--2002-12-
07-Sat-22-00-46-262072/pubring.gpg' fd=3
gpg: DBG: iobuf-1.0: underflow: req=8192
gpg: DBG: iobuf-1.0: underflow: got=963 rc=0
gpg: DBG: parse_packet(iob=1): type=6 length=418 (search.keyring.c.962)
gpg: DBG: mpi_alloc(1024)
gpg: DBG: mpi_alloc_limb_space(1024)
gpg: DBG: mpi_alloc(160)
gpg: DBG: mpi_alloc_limb_space(160)
gpg: DBG: mpi_alloc(1024)
gpg: DBG: mpi_alloc_limb_space(1024)
gpg: DBG: mpi_alloc(1024)
gpg: DBG: mpi_alloc_limb_space(1024)
gpg: DBG: free_packet() type=6
gpg: DBG: mpi_free
gpg: DBG: dummy m_size called
gpg: DBG: mpi_free_limb_space of size 0
gpg: DBG: mpi_free
gpg: DBG: dummy m_size called
gpg: DBG: mpi_free_limb_space of size 0
gpg: DBG: mpi_free
gpg: DBG: dummy m_size called
gpg: DBG: mpi_free_limb_space of size 0
gpg: DBG: mpi_free
gpg: DBG: dummy m_size called
gpg: DBG: mpi_free_limb_space of size 0
gpg: DBG: parse_packet(iob=1): type=13 length=27 (search.keyring.c.962)
gpg: DBG: free_packet() type=13
gpg: DBG: fd_cache_open (/root/tmp--xyzzy-nichi/a-tn-qhome-qout--2002-12-07
-Sat-22-00-46-262072/pubring.gpg) not cached
gpg: DBG: iobuf-2.0: open `/root/tmp--xyzzy-nichi/a-tn-qhome-qout--2002-12-
07-Sat-22-00-46-262072/pubring.gpg' fd=6
gpg: DBG: iobuf-2.0: underflow: req=8192
gpg: DBG: iobuf-2.0: underflow: got=963 rc=0
gpg: DBG: parse_packet(iob=2): type=6 length=418 (parse.keyring.c.382)
gpg: DBG: mpi_alloc(1024)
gpg: DBG: mpi_alloc_limb_space(1024)
gpg: DBG: mpi_alloc(160)
gpg: DBG: mpi_alloc_limb_space(160)
gpg: DBG: mpi_alloc(1024)
gpg: DBG: mpi_alloc_limb_space(1024)
gpg: DBG: mpi_alloc(1024)
gpg: DBG: mpi_alloc_limb_space(1024)
gpg: DBG: parse_packet(iob=2): type=13 length=27 (parse.keyring.c.382)
...
--
judy m. down
<judy0@blackburnmail.com>
--------------------------------------------------------------------
Get your free email address at http://www.blackburnmail.com/
Reply to: