<DKIM> Re:[offlineimap] Duplicates of messages copied from remote to local (#174)

[Peter Lewis]

On Sat, 25 Jul 2015, Janna Martl wrote:
>Probably unrelated, but I'm confused about this "unrelated error" that
>caused you to skip commits: 115d141218c3deb7d2fc92ea87732b4420c4a5d2,
>which you skipped, is one documentation change away from v6.5.7, which
>you marked bad.

I just looked at this and you are right... I came across a couple of
"other" errors while I was bisecting, which I just skipped without
paying too much attention. I do remember there being an error related to
documentation, but that was probably to do with the packaging (I am
installing using Arch's package before testing, for convenience).

Just for all our sanity, I think I will re-do the bisecting and pay more
attention to the errors I am skipping.


But, before that, I will test 3a0565, as you suggest below...

>It would be really great if you could somehow test one of the commits
>that happened after this got reverted and before the rtime-fixing
>patches got in later, just as a sort of sanity check. The only
>candidates for this that you haven't skipped are
>3a056519f9773f96dce936be789c913ad27e321c,
>03d963913c3281301bd96304cfc6950c5f9c23f9, and
>eb90932d4a38b1724aeb1dcbdc0de2ef775f694a.
>
>Otherwise, I don't have any good ideas, so unless Nicholas has some, I
>guess we have to dig in and get more details about what's going wrong.
>Can you try 25513e90387f60831b94c6041de08db5ba12e95f again with
>debugging on (offlineimap -d all) and also with the extra print
>statements I've offered below? Also, can you send this info (minus
>passwords!) for one or two runs before the one that failed (as well as
>for the one that failed).

I will also try this afterwards, but I guess it would be best to be
totally sure about which commit introduced the issue first (in case
there are problems with my bisecting, as above).

Peter.