2009/3/16 Matt Goodall <matt.good...@gmail.com>:
2009/3/16 Matt Goodall <matt.good...@gmail.com>:
2009/3/16 Chris Anderson <jch...@apache.org>:
On Mon, Mar 16, 2009 at 6:07 AM, Jan Lehnardt <j...@apache.org>
wrote:
On 15 Mar 2009, at 21:58, Chris Anderson wrote:
We are within striking distance of 0.9
12 Outstanding issues:
https://issues.apache.org:443/jira/secure/IssueNavigator.jspa?reset=true&&pid=12310780&priority=1&status=1&status=4&fixfor=12313208&sorter/field=issuekey&sorter/order=DESC
I think it'd be very cool if we could release in time for
ApacheCon EU
- starts on March 23rd. I think this is feasible but it will
require
some doing.
Some of these tickets may be as simple as documentation or
exploration
to see if the issue is resolved by the new replication
security patch.
None of them look very challenging.
I see the replication test failing in Firefox and Safari 3.0
(but not Safari
4 beta)
The test fails with
{"error":"normal","reason":"{gen_server,call,
[<0.29400.0>,get_result,infinity]}
which comes as
[info] [<0.29281.0>] 127.0.0.1 - - 'GET'
/test_suite_db_a/_local%2Fb1fc9bd58a179f7aeb55e42d710fdceb 200
[info] [<0.29359.0>] starting replication
"b1fc9bd58a179f7aeb55e42d710fdceb"
at <0.29400.0>
[info] [<0.29400.0>] recording a checkpoint at source
update_seq 20
[error] [<0.29359.0>] Uncaught error in HTTP request: {exit,
{normal,
{gen_server,call,
[<0.29400.0>,get_result,infinity]}}}
in the CouchDB log
I get this intermittently on my FF also. Running it a few times
makes
it pass. I think this is more a problem with the test suite
than with
replication. Can you make this a ticket? I think it'll be a fair
amount of work.
I've seen the same, or very similar, error when triggering a
replication using curl and Futon's replication page. I don't
have any
logs right now though.
- Matt
I've also had problems saving and deleting a document in Futon. The
few times I've had it has been after the document is in conflict
after
replication. Again, no logs at the moment; I'll see if I can repeat
it.
OK, I've got a database (tiny in size) and an error log for this
now.
It looks like it might be related to the replication error Jan
reported. Here's part of the log, it seems to contain the important
stuff:
[error] [<0.2249.0>] Uncaught error in HTTP request: {exit,
{{{case_clause,
{ok,
[{<<"982111398">>,
{false,{4140,9956}},
[{<<"3589267040">>,
{false,{4344,9752}},
[]},
{<<"825225740">>,
{false,{4333,9763}},
[{<<"2045729199">>,
{doc,<<"foo">>,
{7,
[<<"2045729199">>,
<<"825225740">>]},
{[]},
[],true,[]},
[]}]}]}],
false}},
[{couch_key_tree,merge_at,3},
{couch_key_tree,merge_at,3},
{couch_key_tree,merge_one,4},
{couch_key_tree,'-merge/2-fun-0-',
2},
{lists,foldl,3},
{couch_key_tree,merge,2},
{couch_db_updater,
'-merge_rev_trees/6-fun-0-',5},
{lists,foldl,3}]},
{gen_server,call,
[<0.5588.0>,
{update_docs,
[[{doc,<<"foo">>,
{7,
[<<"2045729199">>,<<"825225740">>]},
{[]},
[],true,[]}]],
[]},
infinity]}}}
[info] [<0.2249.0>] 127.0.0.1 - - 'DELETE' /conflicts/foo?
rev=6-825225740 500
- Matt
