Open GoogleCodeExporter opened 9 years ago
I'm not at all surprised... people reported problems with its predecessor
(whatever it was called). It was never a priority for me but since it's built
in now I guess I'm going to have to try to fix it at some point. If anyone has
time to diagnose the specific issue, that will make it more likely I'll get to
it sooner. :)
Original comment by jfitz...@gmail.com
on 3 Oct 2010 at 6:55
I tried to reproduce the bug with way (2) since Sync is too broken to get
anything out of it if I follow (1). I attached the sync-log, maybe that'll help.
Original comment by alex.b...@gmail.com
on 4 Oct 2010 at 6:52
Attachments:
And here's a sync-log from a single sync attempt.
Original comment by alex.b...@gmail.com
on 4 Oct 2010 at 6:53
Attachments:
i think i've identified that FF Sync is not able to store/retrieve it's Sync
Key (previously Secret Phrase) when keychain services addon is installed. so
every browser restart = failed connect and re-entering secret phrase manually.
no other serious breakaeg seems to happen though, so this (FF3.6, Sync 1.5)
should be a relatively calm testing platform.
Original comment by LKooglizmus
on 9 Nov 2010 at 1:33
Original comment by jfitz...@gmail.com
on 12 Nov 2010 at 9:37
Issue 4 has been merged into this issue.
Original comment by jfitz...@gmail.com
on 12 Nov 2010 at 9:42
I've tried
http://code.google.com/p/mozilla-keychain/downloads/detail?name=macos-keychain-1
.1pre2.xpi&can=2&q= now and it still seems to interfere with sync on 4.0b12pre,
albeit not breaking it to the point of unusability. See the sync log,
specifically the very long line beginning with "2011-02-13 10:28:05 Collection
WARN Got exception calling onProgress handler".
Original comment by alex.b...@gmail.com
on 13 Feb 2011 at 9:31
Attachments:
Yes, the new versions are still functionally essentially the same. I had to
rewrite everything using the new js-ctypes FFI interface in order to make
compatibility with 64-bit easier. That should also make it easier to add
support for Keychain "application passwords", which is I think what will be
required to support sync (sync uses a custom URL scheme, which Apple doesn't
support).
I will try to have a look at sync before doing a final release since people are
obviously going to run into this problem with FF 4. The problem is just that my
time is very limited so I'm not sure when exactly I'll be able to get to it.
Original comment by jfitz...@gmail.com
on 16 Feb 2011 at 5:57
Version 1.1 should allow Sync to work by simply shunting its passwords into the
legacy storage mechanism instead of the Keychain. I'll sort out a better fix
sometime but I don't find I have much free time at the moment. The code's
available if anyone wants to take a crack at it.
Original comment by jfitz...@gmail.com
on 23 Mar 2011 at 12:56
Although 1.1 initially seemed to be working, now I'm still getting some error:
1301074479217 Engine.Passwords INFO 0 outgoing items pre-reconciliation
1301074479230 Engine.Passwords WARN Failed to reconcile incoming record
{6605e415-2259-40ea-b978-6bc3138405bb}
1301074479230 Engine.Passwords WARN Encountered exception: 'Error: Not yet
implemented: searchLogins()' when calling method:
[nsILoginManagerStorage::searchLogins] Stack trace:
resource://gre/components/nsLoginManager.js:512 <
PasswordStore__getLoginFromGUID()@resource://services-sync/engines/passwords.js:
157 <
PasswordStore__itemExists()@resource://services-sync/engines/passwords.js:214 <
SyncEngine__reconcile()@resource://services-sync/engines.js:838 <
resource://services-sync/engines.js:659 <
resource://services-sync/record.js:813 <
Channel_onDataAvail()@resource://services-sync/resource.js:530
1301074479231 Collection DEBUG GET success 200
https://myserver/weave/1.0/leho/storage/passwords?full=1&ids={6605e415-2259-40ea
-b978-6bc3138405bb}
1301074479231 Engine.Passwords DEBUG Records that failed to apply:
{6605e415-2259-40ea-b978-6bc3138405bb}
1301074479231 Service.Main DEBUG passwords failed to apply some records.
1301074479232 Engine.Passwords INFO Records: 0 applied, 1 failed to apply, 0
reconciled.
Original comment by LKooglizmus
on 25 Mar 2011 at 5:38
There's Issue 36 already open for the searchLogins() issue but I'm not sure
when I'll find time to fix it. I don't use Sync myself, so I'm afraid it's not
a very high priority for me.
Despite the searchLogins warning, I tested Sync as far as ensuring it was
capable of logging in, so it seemed everything was ok. Nonetheless, I guess I'm
not surprised that it fails at a later point.
Original comment by jfitz...@gmail.com
on 25 Mar 2011 at 8:17
Issue 66 has been merged into this issue.
Original comment by jfitz...@gmail.com
on 5 Aug 2014 at 5:49
Issue 69 has been merged into this issue.
Original comment by jfitz...@gmail.com
on 5 Aug 2014 at 6:02
Original issue reported on code.google.com by
alex.b...@gmail.com
on 2 Oct 2010 at 10:41