[tahoe-dev] [tahoe-lafs] #698: corrupted file displayed to user after failure to download followed by retry
tahoe-lafs
trac at allmydata.org
Thu May 14 14:16:24 PDT 2009
#698: corrupted file displayed to user after failure to download followed by
retry
--------------------------+-------------------------------------------------
Reporter: zooko | Owner:
Type: defect | Status: new
Priority: critical | Milestone: 1.5.0
Component: code-network | Version: 1.4.1
Keywords: integrity | Launchpad_bug:
--------------------------+-------------------------------------------------
Comment(by zooko):
By the way, the client doing the fetch was Firefox 3.5 beta 4 -- more
details in the original report. It's the package named "firefox-3.5" in
Ubuntu Jaunty.
Also, Brian, what do you think about the fact that there was no download
step after the
{{{
09:47:10 07-May-2009 retrieve 6s64wyhfbm7yxb5cwzqblnndpe No
2.0kB 100.0% Done
09:47:10 07-May-2009 mapupdate MODE_READ 6s64wyhfbm7yxb5cwzqblnndpe
No -NA- 100.0% Done
}}}
When I reloaded, but there was a download step after those two steps when
I initially loaded? This means the client (Firefox-3.5b4 web browser)
didn't actually fetch the contents of my blog when I hit "Reload" and
instead just spewed that messed-up page (attached as "wiki.html.bz2"),
right? In that case the "displaying corrupt page" is actually Firefox's
problem (although as David-Sarah Hopwood pointed out
[http://allmydata.org/pipermail/tahoe-dev/2009-May/001724.html on the
mailing list], the way that Tahoe commits to a 200 success code before
completely downloading the file means this can always happen by accident).
Certainly ''something'' that Tahoe did was unusual, see the attached file
{{{incident-2009-05-07-094319-jg54cni.flog.bz2}}}, which, when viewed with
{{{flogtool web-viewer incident-2009-05-07-094319-jg54cni.flog.bz2}}}
shows this among other things:
{{{
# Incident Triggers:
* 09:43:19.240 [31906]: SCARY <CiphertextDownloader
#14>(lxershd2xflh): download failed! FAILURE:
[CopiedFailure instance: Traceback from remote host -- Traceback
(most recent call last):
Failure: twisted.internet.defer.FirstError:
FirstError(<twisted.python.failure.Failure <class
'foolscap.ipb.DeadReferenceError'>>, 2)
]
[INCIDENT-TRIGGER]
}}}
--
Ticket URL: <http://allmydata.org/trac/tahoe/ticket/698#comment:2>
tahoe-lafs <http://allmydata.org>
secure decentralized file storage grid
More information about the tahoe-dev
mailing list