tahoe-lafs
/
trac-2024-07-25
2
0
Issues 1.1k Wiki

corrupted file displayed to user after failure to download followed by retry #698

New Issue
Closed
opened 2009-05-07 16:21:42 +00:00 by zooko · 19 comments
zooko commented 2009-05-07 16:21:42 +00:00
Copy Link

I clicked the bookmark to load my blog writably, and I got an HTML page saying:

<class 'twisted.internet.defer.FirstError'>: FirstError(<twisted.python.failure.Failure <class 'foolscap.ipb.DeadReferenceError'>>, 2)

<class 'twisted.internet.defer.FirstError'>: FirstError(<twisted.python.failure.Failure <class 'foolscap.ipb.DeadReferenceError'>>, 2)

I looked at the "Recent Uploads/Downloads" page and saw that my attempt to load it had failed:

09:23:12 07-May-2009  	download  	lxershd2xflho66w6yikhwg3ne  	No  	588.7kB  	0.0%  	Failed
09:23:12 07-May-2009 	retrieve 	6s64wyhfbm7yxb5cwzqblnndpe 	No 	2.0kB 	100.0% 	Done
09:23:12 07-May-2009 	mapupdate MODE_READ 	6s64wyhfbm7yxb5cwzqblnndpe 	No 	-NA- 	100.0% 	Done

The three details pages are attached: mapupdate-35.html, retrieve-35.html, and down-13.html.

I looked in the logs directory of my Tahoe node, and saw that the twistd.log had these same error messages about DeadReferenceError. twistd.log is attached (bzipped).

I looked in the logs/incidents directory and saw that there was one incident that was recorded at the time of this attempt to load. It is attached as incident-2009-05-07-094319-jg54cni.flog.bz2. The triggering incident line is

13:51:05.704 [6928]: SCARY <CiphertextDownloader #10>(hekksgbfsn6w): 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'>>, 0)
]

So far I think that this Tahoe demonstrating suboptimal handling of a network failure -- it should probably have returned an HTTP 503 "Service Unavailable" (or maybe 504 "Gateway Timeout" or just 500 "Internal Server Error"?) instead of an HTML page containing cryptic error messages. But it gets worse:

Then I hit the "Reload" button on my web browser, and I got the same two error message lines followed by a partial copy of the contents of my blog source code! This result is attached as wiki.html (bzipped). This is what I mean by a corrupted file being displayed to the user.

The "Recent Uploads and Downloads" now says:

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
09:23:12 07-May-2009 	download 	lxershd2xflho66w6yikhwg3ne 	No 	588.7kB 	0.0% 	Failed
09:23:12 07-May-2009 	retrieve 	6s64wyhfbm7yxb5cwzqblnndpe 	No 	2.0kB 	100.0% 	Done
09:23:12 07-May-2009 	mapupdate MODE_READ 	6s64wyhfbm7yxb5cwzqblnndpe 	No 	-NA- 	100.0% 	Done

The fact that there is no download following the map-update is surprising to me.

The details from the new mapupdate-36.html and retrieve-36.html are attached. There are no new problems reported in the twistd.log or the logs/incidents/.

I'm going to go ahead and mark this with Priority: critical because I see a corrupted file and I don't understand why. Hopefully it will turn out to be a bug in the web browser, which is an unstable release: firefox-3.5 3.5~b4~hg20090330r24021+nobinonly-0ubuntu1.

I clicked the bookmark to load my blog writably, and I got an HTML page saying: ``` <class 'twisted.internet.defer.FirstError'>: FirstError(<twisted.python.failure.Failure <class 'foolscap.ipb.DeadReferenceError'>>, 2) <class 'twisted.internet.defer.FirstError'>: FirstError(<twisted.python.failure.Failure <class 'foolscap.ipb.DeadReferenceError'>>, 2) ``` I looked at the "Recent Uploads/Downloads" page and saw that my attempt to load it had failed: ``` 09:23:12 07-May-2009 download lxershd2xflho66w6yikhwg3ne No 588.7kB 0.0% Failed 09:23:12 07-May-2009 retrieve 6s64wyhfbm7yxb5cwzqblnndpe No 2.0kB 100.0% Done 09:23:12 07-May-2009 mapupdate MODE_READ 6s64wyhfbm7yxb5cwzqblnndpe No -NA- 100.0% Done ``` The three details pages are attached: `mapupdate-35.html`, `retrieve-35.html`, and `down-13.html`. I looked in the `logs` directory of my Tahoe node, and saw that the `twistd.log` had these same error messages about `DeadReferenceError`. `twistd.log` is attached (bzipped). I looked in the `logs/incidents` directory and saw that there was one incident that was recorded at the time of this attempt to load. It is attached as `incident-2009-05-07-094319-jg54cni.flog.bz2`. The triggering incident line is ``` 13:51:05.704 [6928]: SCARY <CiphertextDownloader #10>(hekksgbfsn6w): 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'>>, 0) ] ``` So far I think that this Tahoe demonstrating suboptimal handling of a network failure -- it should probably have returned an HTTP 503 "Service Unavailable" (or maybe 504 "Gateway Timeout" or just 500 "Internal Server Error"?) instead of an HTML page containing cryptic error messages. But it gets worse: Then I hit the "Reload" button on my web browser, and I got the same two error message lines followed by a partial copy of the contents of my blog source code! This result is attached as `wiki.html` (bzipped). This is what I mean by a corrupted file being displayed to the user. The "Recent Uploads and Downloads" now says: ``` 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 09:23:12 07-May-2009 download lxershd2xflho66w6yikhwg3ne No 588.7kB 0.0% Failed 09:23:12 07-May-2009 retrieve 6s64wyhfbm7yxb5cwzqblnndpe No 2.0kB 100.0% Done 09:23:12 07-May-2009 mapupdate MODE_READ 6s64wyhfbm7yxb5cwzqblnndpe No -NA- 100.0% Done ``` The fact that there is no download following the map-update is surprising to me. The details from the new `mapupdate-36.html` and `retrieve-36.html` are attached. There are no new problems reported in the `twistd.log` or the `logs/incidents/`. I'm going to go ahead and mark this with `Priority: critical` because I see a corrupted file and I don't understand why. Hopefully it will turn out to be a bug in the web browser, which is an unstable release: firefox-3.5 `3.5~b4~hg20090330r24021+nobinonly-0ubuntu1`.
zooko added the
code-network
critical
defect
1.4.1
 labels 2009-05-07 16:21:42 +00:00
zooko added this to the 1.5.0 milestone 2009-05-07 16:21:42 +00:00
zooko commented 2009-05-07 16:22:51 +00:00
Author
Copy Link

Attachment mapupdate-35.html (1745 bytes) added

**Attachment** mapupdate-35.html (1745 bytes) added
mapupdate-35.html
1.7 KiB
zooko commented 2009-05-07 16:23:00 +00:00
Author
Copy Link

Attachment retrieve-35.html (1000 bytes) added

**Attachment** retrieve-35.html (1000 bytes) added
retrieve-35.html
1000 B
zooko commented 2009-05-07 16:23:07 +00:00
Author
Copy Link

Attachment down-13.html (1817 bytes) added

**Attachment** down-13.html (1817 bytes) added
down-13.html
1.8 KiB
zooko commented 2009-05-07 16:23:55 +00:00
Author
Copy Link

Attachment twistd.log.bz2 (7368 bytes) added

**Attachment** twistd.log.bz2 (7368 bytes) added
twistd.log.bz2
7.2 KiB
zooko commented 2009-05-07 16:24:45 +00:00
Author
Copy Link

Attachment incident-2009-05-07-094319-jg54cni.flog.bz2 (38154 bytes) added

**Attachment** incident-2009-05-07-094319-jg54cni.flog.bz2 (38154 bytes) added
incident-2009-05-07-094319-jg54cni.flog.bz2
37 KiB
zooko commented 2009-05-07 16:25:33 +00:00
Author
Copy Link

Attachment wiki.html.bz2 (147850 bytes) added

**Attachment** wiki.html.bz2 (147850 bytes) added
wiki.html.bz2
144 KiB
zooko commented 2009-05-07 16:25:52 +00:00
Author
Copy Link

Attachment mapupdate-36.html (1744 bytes) added

**Attachment** mapupdate-36.html (1744 bytes) added
mapupdate-36.html
1.7 KiB
zooko commented 2009-05-07 16:26:01 +00:00
Author
Copy Link

Attachment retrieve-36.html (1000 bytes) added

**Attachment** retrieve-36.html (1000 bytes) added
retrieve-36.html
1000 B
warner commented 2009-05-07 19:47:52 +00:00
Copy Link

That's pretty weird. The first thing that comes to mind is that a server
connection could have been lost in the middle of the download (in this case,
after we've retrieved the UEB and some of the hashes, but before we've
retrieved the first data block). The web server has to commit to success
(200) or failure (404 or 500 or something) before it starts sending any of
the plaintext, but it doesn't want to store the entire file either. So it
bases the HTTP response code upon the initial availability of k servers, and
hopes they'll stick around for the whole download.

When we get a "late failure" (i.e. one of the servers disconnects in the
middle), the webapi doesn't have a lot of choices. At the moment, it emits a
brief error message (attached to whatever partial content has already been
written out), then drops the HTTP connection, and hopes that the client is
observant enough to notice that the number of received bytes does not match
the previously-sent Content-Length header, and then announce an error on the
client side.

If the application doing the fetch (perhaps the browser, perhaps tiddywiki
itself?) doesn't strictly check the Content-Length header, then it could get
partial content without an error message.

There are two directions to fix this:

  • change the webapi to use "Chunked Encoding", basically delivering data one
    segment at a time, possibly giving the server a chance to emit an error
    header in between segments: this would let us respond better to these
    errors
  • fix the other download-should-be-better tickets (#193, #287) to tolerate
    lost servers better, which might reduce the rate at which these errors
    occur

I'm not sure what's up with the happens-again-after-retry part of this. For
the benefit of partial-range fetches, we sometimes cache the file's contents
locally, and I don't know how that would interact with lost-server errors.
It's at least conceivable that the caching mechanism doesn't realize that an
error occurred, and tries to pass partial data to the second download
attempt. But most browsers don't send a Range: header at all (it's mostly
streaming media players which do that), and I believe that the webapi will
skip this whole caching thing unless it sees a Range: header.

That's pretty weird. The first thing that comes to mind is that a server connection could have been lost in the middle of the download (in this case, after we've retrieved the UEB and some of the hashes, but before we've retrieved the first data block). The web server has to commit to success (200) or failure (404 or 500 or something) before it starts sending any of the plaintext, but it doesn't want to store the entire file either. So it bases the HTTP response code upon the initial availability of k servers, and hopes they'll stick around for the whole download. When we get a "late failure" (i.e. one of the servers disconnects in the middle), the webapi doesn't have a lot of choices. At the moment, it emits a brief error message (attached to whatever partial content has already been written out), then drops the HTTP connection, and hopes that the client is observant enough to notice that the number of received bytes does not match the previously-sent Content-Length header, and then announce an error on the client side. If the application doing the fetch (perhaps the browser, perhaps tiddywiki itself?) doesn't strictly check the Content-Length header, then it could get partial content without an error message. There are two directions to fix this: * change the webapi to use "Chunked Encoding", basically delivering data one segment at a time, possibly giving the server a chance to emit an error header in between segments: this would let us respond better to these errors * fix the other download-should-be-better tickets (#193, #287) to tolerate lost servers better, which might reduce the rate at which these errors occur I'm not sure what's up with the happens-again-after-retry part of this. For the benefit of partial-range fetches, we sometimes cache the file's contents locally, and I don't know how that would interact with lost-server errors. It's at least conceivable that the caching mechanism doesn't realize that an error occurred, and tries to pass partial data to the second download attempt. But most browsers don't send a Range: header at all (it's mostly streaming media players which do that), and I believe that the webapi will skip this whole caching thing unless it sees a Range: header.
zooko commented 2009-05-14 21:16:24 +00:00
Author
Copy Link

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 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]
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 [on the mailing list](http://allmydata.org/pipermail/tahoe-dev/2009-May/001724.html), 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] ```
zooko commented 2009-05-14 21:45:09 +00:00
Author
Copy Link

I just got this failure again. The file in question (my blog) has been appended-to since the original report, so now its total size is different from last time. Here is the accompanying foolscap incident report. I did not hit re-load in this case, I just opened a new tab and clicked on the bookmark that opens up my blog in writeable mode.

I just got this failure again. The file in question (my blog) has been appended-to since the original report, so now its total size is different from last time. Here is the accompanying foolscap incident report. I did *not* hit re-load in this case, I just opened a new tab and clicked on the bookmark that opens up my blog in writeable mode.
zooko commented 2009-05-14 21:46:13 +00:00
Author
Copy Link

Attachment incident-2009-05-14-142935-v7cl57a.flog.bz2 (14926 bytes) added

**Attachment** incident-2009-05-14-142935-v7cl57a.flog.bz2 (14926 bytes) added
incident-2009-05-14-142935-v7cl57a.flog.bz2
15 KiB
zooko commented 2009-05-14 21:55:27 +00:00
Author
Copy Link

The triggering incident in that new incident log (as displayed by flogtool web-viewer is:

# Incident Triggers:

    * 14:29:35.644 [816]: WEIRD Tub.connectorFinished: WEIRD, <foolscap.negotiate.TubConnector object at 0x21ace10 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to qvqv7jmm76yfhdjuuzeqyfuwlqinpofd> is not in [<foolscap.negotiate.TubConnector object at 0x20a8510 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to gdfa2zqvj7l2ng26bwalyubitdxtmywf>, <foolscap.negotiate.TubConnector object at 0x20a8810 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to i3u2uz4ewpz3n36ckaqmtmfrubpifapd>, <foolscap.negotiate.TubConnector object at 0x20b3450 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to x4pds7hcgaq2exnumxbsjkv2jrkzyyxf>, <foolscap.negotiate.TubConnector object at 0x20b3750 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to k6h4otc2f2wmmysam5ofciaenhngurwm>, <foolscap.negotiate.TubConnector object at 0x20b3d50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to gkc4vgeeyktdzwklygxnummy3cin2fri>, <foolscap.negotiate.TubConnector object at 0x20bb090 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to srb3paczogfsh6blsmopelry2ogc4uii>, <foolscap.negotiate.TubConnector object at 0x20bb410 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to jg6qoqzfunqhsvjsrquize3tpcsyglvo>, <foolscap.negotiate.TubConnector object at 0x20bbd90 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to pstqfjqxsfpu7ac7crfzgm26jervfz37>, <foolscap.negotiate.TubConnector object at 0x20c5090 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to mtuj2lnh5sarkpummx77w7jjhynz2jea>, <foolscap.negotiate.TubConnector object at 0x20c5350 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to okjeolejndkuenpuuaiduvk2plebgjao>, <foolscap.negotiate.TubConnector object at 0x20c5710 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to jfccf453ml75gscmpugihlg2bcjzgswr>, <foolscap.negotiate.TubConnector object at 0x20c5c10 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to yzdb43kciqbv3oech66pq54lrtddbbvf>, <foolscap.negotiate.TubConnector object at 0x20d0250 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to i4kd2dpoaj7wnd6kicsklcswayop24wy>, <foolscap.negotiate.TubConnector object at 0x20d0a90 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to jobn3i5i7dlu6vt2y6duc3yhvif3d4vr>, <foolscap.negotiate.TubConnector object at 0x20d0f50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to zzkq2nfnpukh5uajj7aedsl6sc3m5rup>, <foolscap.negotiate.TubConnector object at 0x20d9d50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to t5zibyamutxhjtazj7fdc5q4nqmifqnj>, <foolscap.negotiate.TubConnector object at 0x21ac150 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to sttpea3ze6sqi5qyqmomeiwfsayytkar>, <foolscap.negotiate.TubConnector object at 0x21b8210 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to 3wrurfqxq4xqzlqatks53b2ekwaw5jt2>, <foolscap.negotiate.TubConnector object at 0x21b8850 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to g5662mtbvxlfcipvcpvicellm2ljdqow>, <foolscap.negotiate.TubConnector object at 0x21bf350 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to 6fyx5u4zr7tvz3szynihc4x3uc6ct5gh>, <foolscap.negotiate.TubConnector object at 0x21bf650 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to xpyajfs4rvm3yps5szojjrxjsgkmshfp>, <foolscap.negotiate.TubConnector object at 0x21bfb50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to lwkv6cjicbzqjwwwuifik3pogeupsicb>, <foolscap.negotiate.TubConnector object at 0x21ca190 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to xiktf6ok5f5ao5znxxttriv233hmvi4v>, <foolscap.negotiate.TubConnector object at 0x21ca490 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to t5g7egomnnktbpydbuijt6zgtmw4oqi5>, <foolscap.negotiate.TubConnector object at 0x21caa90 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to odcd5enlzmv7iobwad63tchotty3lotx>, <foolscap.negotiate.TubConnector object at 0x21e0710 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to iul3ljbknqud666muthssyfbyrbd4xae>] [INCIDENT-TRIGGER]
The triggering incident in that new incident log (as displayed by `flogtool web-viewer` is: ``` # Incident Triggers: * 14:29:35.644 [816]: WEIRD Tub.connectorFinished: WEIRD, <foolscap.negotiate.TubConnector object at 0x21ace10 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to qvqv7jmm76yfhdjuuzeqyfuwlqinpofd> is not in [<foolscap.negotiate.TubConnector object at 0x20a8510 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to gdfa2zqvj7l2ng26bwalyubitdxtmywf>, <foolscap.negotiate.TubConnector object at 0x20a8810 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to i3u2uz4ewpz3n36ckaqmtmfrubpifapd>, <foolscap.negotiate.TubConnector object at 0x20b3450 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to x4pds7hcgaq2exnumxbsjkv2jrkzyyxf>, <foolscap.negotiate.TubConnector object at 0x20b3750 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to k6h4otc2f2wmmysam5ofciaenhngurwm>, <foolscap.negotiate.TubConnector object at 0x20b3d50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to gkc4vgeeyktdzwklygxnummy3cin2fri>, <foolscap.negotiate.TubConnector object at 0x20bb090 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to srb3paczogfsh6blsmopelry2ogc4uii>, <foolscap.negotiate.TubConnector object at 0x20bb410 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to jg6qoqzfunqhsvjsrquize3tpcsyglvo>, <foolscap.negotiate.TubConnector object at 0x20bbd90 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to pstqfjqxsfpu7ac7crfzgm26jervfz37>, <foolscap.negotiate.TubConnector object at 0x20c5090 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to mtuj2lnh5sarkpummx77w7jjhynz2jea>, <foolscap.negotiate.TubConnector object at 0x20c5350 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to okjeolejndkuenpuuaiduvk2plebgjao>, <foolscap.negotiate.TubConnector object at 0x20c5710 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to jfccf453ml75gscmpugihlg2bcjzgswr>, <foolscap.negotiate.TubConnector object at 0x20c5c10 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to yzdb43kciqbv3oech66pq54lrtddbbvf>, <foolscap.negotiate.TubConnector object at 0x20d0250 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to i4kd2dpoaj7wnd6kicsklcswayop24wy>, <foolscap.negotiate.TubConnector object at 0x20d0a90 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to jobn3i5i7dlu6vt2y6duc3yhvif3d4vr>, <foolscap.negotiate.TubConnector object at 0x20d0f50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to zzkq2nfnpukh5uajj7aedsl6sc3m5rup>, <foolscap.negotiate.TubConnector object at 0x20d9d50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to t5zibyamutxhjtazj7fdc5q4nqmifqnj>, <foolscap.negotiate.TubConnector object at 0x21ac150 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to sttpea3ze6sqi5qyqmomeiwfsayytkar>, <foolscap.negotiate.TubConnector object at 0x21b8210 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to 3wrurfqxq4xqzlqatks53b2ekwaw5jt2>, <foolscap.negotiate.TubConnector object at 0x21b8850 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to g5662mtbvxlfcipvcpvicellm2ljdqow>, <foolscap.negotiate.TubConnector object at 0x21bf350 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to 6fyx5u4zr7tvz3szynihc4x3uc6ct5gh>, <foolscap.negotiate.TubConnector object at 0x21bf650 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to xpyajfs4rvm3yps5szojjrxjsgkmshfp>, <foolscap.negotiate.TubConnector object at 0x21bfb50 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to lwkv6cjicbzqjwwwuifik3pogeupsicb>, <foolscap.negotiate.TubConnector object at 0x21ca190 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to xiktf6ok5f5ao5znxxttriv233hmvi4v>, <foolscap.negotiate.TubConnector object at 0x21ca490 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to t5g7egomnnktbpydbuijt6zgtmw4oqi5>, <foolscap.negotiate.TubConnector object at 0x21caa90 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to odcd5enlzmv7iobwad63tchotty3lotx>, <foolscap.negotiate.TubConnector object at 0x21e0710 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to iul3ljbknqud666muthssyfbyrbd4xae>] [INCIDENT-TRIGGER] ```
zooko commented 2009-05-14 22:01:42 +00:00
Author
Copy Link

So I think we need to split this ticket up into (at least):

  1. A ticket to switch the wui/wapi to chunked encoding so that we can unambiguously signal late errors to the client.

  2. This ticket, whose focus is probably going to turn to seeing if I can reproduce this problem with any other web browser than the Firefox-3.5b4 that I've seen it on.

  3. A ticket to investigate the WEIRD Tub.connectorFinished mentioned in comment:70966. Is that right, Brian, that it deserves investigation?

So I think we need to split this ticket up into (at least): 1. A ticket to switch the wui/wapi to chunked encoding so that we can unambiguously signal late errors to the client. 2. This ticket, whose focus is probably going to turn to seeing if I can reproduce this problem with any other web browser than the Firefox-3.5b4 that I've seen it on. 3. A ticket to investigate the `WEIRD Tub.connectorFinished` mentioned in [comment:70966](/tahoe-lafs/trac-2024-07-25/issues/698#issuecomment-70966). Is that right, Brian, that it deserves investigation?
zooko commented 2009-05-14 22:33:58 +00:00
Author
Copy Link

Okay, re topic 2 from comment:70967, I see that the corrupted page that I get is actually the complete contents of my wiki with the first 2302 or 2306 bytes of it overwritten by the Twisted error messages.

That is: my wiki begins with the following 4000 bytes:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xml:lang="en" xmlns="http://www.w3.org/1999/xhtml" lang="en"><head>


<script id="versionArea" type="text/javascript">
//<![CDATA[
var version = {title: "TiddlyWiki", major: 2, minor: 5, revision: 0, date: new Date("Mar 9, 2009"), extensions: {}};
//]]>
</script>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta name="copyright" content="TiddlyWiki created by Jeremy Ruston, (jeremy [at] osmosoft [dot] com)

Copyright (c) UnaMesa Association 2004-2009

Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.

Redistributions in binary form must reproduce the above copyright notice, this
list of conditions and the following disclaimer in the documentation and/or other
materials provided with the distribution.

Neither the name of the UnaMesa Association nor the names of its contributors may be
used to endorse or promote products derived from this software without specific
prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 'AS IS' AND ANY
EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
DAMAGE.">
<script id="jsheadArea" type="text/javascript">
//<![CDATA[
/*
 * jQuery JavaScript Library v1.3.2
 * http://jquery.com/
 *
 * Copyright (c) 2009 John Resig
 * Dual licensed under the MIT and GPL licenses.
 * http://docs.jquery.com/License
 *
 * Date: 2009-02-19 17:34:21 -0500 (Thu, 19 Feb 2009)
 * Revision: 6246
 */
(function(){var l=this,g,y=l.jQuery,p=l.$,o=l.jQuery=l.$=function(E,F){return new o.fn.init(E,F)},D=/^[^<]*(<(.|\s)+>)[^>]*$|^#([\w-]+)$/,f=/^.[^:#\[\.,]*$/;o.fn=o.prototype={init:function(E,H){E=E||document;if(E.nodeType){this[0]=E;this.length=1;this.context=E;return this}if(typeof E==="string"){var G=D.exec(E);if(G&&(G[1]||!H)){if(G[1]){E=o.clean([G[1]],H)}else{var I=document.getElementById(G[3]);if(I&&I.id!=G[3]){return o().find(E)}var F=o(I||[]);F.context=document;F.selector=E;return F}}else{return o(H).find(E)}}else{if(o.isFunction(E)){return o(document).ready(E)}}if(E.selector&&E.context){this.selector=E.selector;this.context=E.context}return this.setArray(o.isArray(E)?E:o.makeArray(E))},selector:"",jquery:"1.3.2",size:function(){return this.length},get:function(E){return E===g?Array.prototype.slice.call(this):this[E]},pushStack:function(F,H,E){var G=o(F);G.prevObject=this;G.context=this.context;if(H==="find"){G.selector=this.selector+(this.selector?" ":"")+E}else{if(H){G.selector=this.selector+"."+H+"("+E+")"}}return G},setArray:function(E){this.length=0;Array.prototype.push.apply(this,E);return this},each:function(F,E){return o.each(this,F,E)},index:function(E){return o.inArray(E&&E.jquery?E[0]:E,this)},attr:function(F,H,G){var E=F;if(typeof F==="string"){if(H===g){return this[0]&&o[G||"attr"](this[0],F)}else{E={};E[F]=H}}return this.each(function(I){for(F in E){o.attr(G?this.style:this,F,o.prop(this,E[F],G,I,F))}})},css:function(E,F){if((E=="width"||E=="height")&&parseFloat(F)<0){F=g}return this.attr(E,F,"curCSS")},text:function(F){if(typeof F!=="object"&&F!=null){return this.empty().append((this[0]&&this[0].ownerDocument||do

When I encountered this failure the first time, the first 4000 bytes of the resulting page that I got looked like this:

<html><head><title>Exception</title></head><body><style type="text/css">
p.error {
  color: black;
  font-family: Verdana, Arial, helvetica, sans-serif;
  font-weight: bold;
  font-size: large;
  margin: 0.25em;
}

div {
  font-family: Verdana, Arial, helvetica, sans-serif;
}

strong.variableClass {
  font-size: small;
}

div.stackTrace {
}

div.frame {
  padding: 0.25em;
  background: white;
  border-bottom: thin black dotted;
}

div.firstFrame {
  padding: 0.25em;
  background: white;
  border-top: thin black dotted;
  border-bottom: thin black dotted;
}

div.location {
    font-size: small;
}

div.snippet {
  background: #FFFFDD;
  padding: 0.25em;
}

div.snippetHighlightLine {
  color: red;
}

span.lineno {
    font-size: small;
}

pre.code {
  margin: 0px;
  padding: 0px;
  display: inline;
  font-size: small;
  font-family: "Courier New", courier, monotype;
}

span.function {
  font-weight: bold;
  font-family: "Courier New", courier, monotype;
}

table.variables {
  border-collapse: collapse;
  width: 100%;
}

td.varName {
  width: 1in;
  vertical-align: top;
  font-style: italic;
  font-size: small;
  padding-right: 0.25em;
}

td.varValue {
  padding-left: 0.25em;
  padding-right: 0.25em;
  font-size: small;
}

div.variables {
  margin-top: 0.5em;
}

div.dict {
  background: #cccc99;
  padding: 2px;
  float: left;
}

td.dictKey {
  background: #ffff99;
  font-weight: bold;
}

td.dictValue {
  background: #ffff99;
}

div.list {
  background: #7777cc;
  padding: 2px;
  float: left;
}

div.listItem {
  background: #9999ff;
}

div.instance {
  width: 100%;
  background: #efefef;
  padding: 2px;
  float: left;
}

span.instanceName {
  font-size: small;
  display: block;
}

span.instanceRepr {
  font-family: "Courier New", courier, monotype;
}

div.function {
  background: orange;
  font-weight: bold;
  float: left;
}
</style><a href="#tracebackEnd"><p class="error">&lt;class 'twisted.internet.defer.FirstError'&gt;: FirstError(&lt;twisted.python.failure.Failure &lt;class 'foolscap.ipb.DeadReferenceError'&gt;&gt;, 2)</p></a><div class="stackTrace"></div><a name="tracebackEnd"><p class="error">&lt;class 'twisted.internet.defer.FirstError'&gt;: FirstError(&lt;twisted.python.failure.Failure &lt;class 'foolscap.ipb.DeadReferenceError'&gt;&gt;, 2)</p></a></body></html>, 19 Feb 2009)
 * Revision: 6246
 */
(function(){var l=this,g,y=l.jQuery,p=l.$,o=l.jQuery=l.$=function(E,F){return new o.fn.init(E,F)},D=/^[^<]*(<(.|\s)+>)[^>]*$|^#([\w-]+)$/,f=/^.[^:#\[\.,]*$/;o.fn=o.prototype={init:function(E,H){E=E||document;if(E.nodeType){this[0]=E;this.length=1;this.context=E;return this}if(typeof E==="string"){var G=D.exec(E);if(G&&(G[1]||!H)){if(G[1]){E=o.clean([G[1]],H)}else{var I=document.getElementById(G[3]);if(I&&I.id!=G[3]){return o().find(E)}var F=o(I||[]);F.context=document;F.selector=E;return F}}else{return o(H).find(E)}}else{if(o.isFunction(E)){return o(document).ready(E)}}if(E.selector&&E.context){this.selector=E.selector;this.context=E.context}return this.setArray(o.isArray(E)?E:o.makeArray(E))},selector:"",jquery:"1.3.2",size:function(){return this.length},get:function(E){return E===g?Array.prototype.slice.call(this):this[E]},pushStack:function(F,H,E){var G=o(F);G.prevObject=this;G.context=this.context;if(H==="find"){G.selector=this.selector+(this.selector?" ":"")+E}else{if(H){G.selector=this.selector+"."+H+"("+E+")"}}return G},setArray:function(E){this.length=0;Array.prototype.push.apply(this,E);return this},each:function(F,E){return o.each(this,F,E)},index:function(E){return o.inArray(E&&E.jquery?E[0]:E,this)},attr:function(F,H,G){var E=F;if(typeof F==="string"){if(H===g){return this[0]&&o[G||"attr"](this[0],F)}else{E={};E[F]=H}}return this.each(function(I){for(F in E){o.attr(G?this.style:this,F,o.prop(this,E[F],G,I,F))}})},css:function(E,F){if((E=="width"||E=="height")&&parseFloat(F)<0){F=g}return this.attr(E,F,"curCSS")},text:function(F){if(typeof F!=="object"&&F!=null){return this.empty().append((this[0]&&this[0].ownerDocument

See what I mean? It looks like Firefox 3.5b4 has overwritten the first 2302 or so bytes of the file with the twisted error message. Note that it doesn't exactly line up -- the twisted error message from <html> to </html> inclusive is 2302 bytes, but the first bytes of the original file , 19 Feb 2009) I would expect to begin on byte 2306 rather than byte 2302. Maybe there was a newline conversion or the header got subtly rewritten a little between the time I captured the above view of "my wiki" and the time the bad wiki was emitted.

Anyway, this makes me pretty sure that Firefox-3.5b4 is doing something mighty funny here. Certainly I don't think Tahoe is sending those bytes down in an HTTP response. Indeed, from the "Current Uploads Downloads" log, we can see that Tahoe wasn't asked for the contents of the wiki at all (the "update in MODE_READ" and the "retrieve" operations are the get the directory, not to get the actual wiki.html file from the directory). Firefox-3.5b4 must have supplied the cached contents of the wiki itself, then overwritten the first few bytes of it with the twisted error message, then displayed it to the user. Ugh.

I suppose we should report this on the firefox issue tracker...

Okay, re topic 2 from [comment:70967](/tahoe-lafs/trac-2024-07-25/issues/698#issuecomment-70967), I see that the corrupted page that I get is actually the *complete* contents of my wiki *with the first 2302 or 2306 bytes of it overwritten by the Twisted error messages*. That is: my wiki begins with the following 4000 bytes: ``` <!DOCTYPE HTML PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html xml:lang="en" xmlns="http://www.w3.org/1999/xhtml" lang="en"><head> <script id="versionArea" type="text/javascript"> //<![CDATA[ var version = {title: "TiddlyWiki", major: 2, minor: 5, revision: 0, date: new Date("Mar 9, 2009"), extensions: {}}; //]]> </script> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> <meta name="copyright" content="TiddlyWiki created by Jeremy Ruston, (jeremy [at] osmosoft [dot] com) Copyright (c) UnaMesa Association 2004-2009 Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. Neither the name of the UnaMesa Association nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 'AS IS' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE."> <script id="jsheadArea" type="text/javascript"> //<![CDATA[ /* * jQuery JavaScript Library v1.3.2 * http://jquery.com/ * * Copyright (c) 2009 John Resig * Dual licensed under the MIT and GPL licenses. * http://docs.jquery.com/License * * Date: 2009-02-19 17:34:21 -0500 (Thu, 19 Feb 2009) * Revision: 6246 */ (function(){var l=this,g,y=l.jQuery,p=l.$,o=l.jQuery=l.$=function(E,F){return new o.fn.init(E,F)},D=/^[^<]*(<(.|\s)+>)[^>]*$|^#([\w-]+)$/,f=/^.[^:#\[\.,]*$/;o.fn=o.prototype={init:function(E,H){E=E||document;if(E.nodeType){this[0]=E;this.length=1;this.context=E;return this}if(typeof E==="string"){var G=D.exec(E);if(G&&(G[1]||!H)){if(G[1]){E=o.clean([G[1]],H)}else{var I=document.getElementById(G[3]);if(I&&I.id!=G[3]){return o().find(E)}var F=o(I||[]);F.context=document;F.selector=E;return F}}else{return o(H).find(E)}}else{if(o.isFunction(E)){return o(document).ready(E)}}if(E.selector&&E.context){this.selector=E.selector;this.context=E.context}return this.setArray(o.isArray(E)?E:o.makeArray(E))},selector:"",jquery:"1.3.2",size:function(){return this.length},get:function(E){return E===g?Array.prototype.slice.call(this):this[E]},pushStack:function(F,H,E){var G=o(F);G.prevObject=this;G.context=this.context;if(H==="find"){G.selector=this.selector+(this.selector?" ":"")+E}else{if(H){G.selector=this.selector+"."+H+"("+E+")"}}return G},setArray:function(E){this.length=0;Array.prototype.push.apply(this,E);return this},each:function(F,E){return o.each(this,F,E)},index:function(E){return o.inArray(E&&E.jquery?E[0]:E,this)},attr:function(F,H,G){var E=F;if(typeof F==="string"){if(H===g){return this[0]&&o[G||"attr"](this[0],F)}else{E={};E[F]=H}}return this.each(function(I){for(F in E){o.attr(G?this.style:this,F,o.prop(this,E[F],G,I,F))}})},css:function(E,F){if((E=="width"||E=="height")&&parseFloat(F)<0){F=g}return this.attr(E,F,"curCSS")},text:function(F){if(typeof F!=="object"&&F!=null){return this.empty().append((this[0]&&this[0].ownerDocument||do ``` When I encountered this failure the first time, the first 4000 bytes of the resulting page that I got looked like this: ``` <html><head><title>Exception</title></head><body><style type="text/css"> p.error { color: black; font-family: Verdana, Arial, helvetica, sans-serif; font-weight: bold; font-size: large; margin: 0.25em; } div { font-family: Verdana, Arial, helvetica, sans-serif; } strong.variableClass { font-size: small; } div.stackTrace { } div.frame { padding: 0.25em; background: white; border-bottom: thin black dotted; } div.firstFrame { padding: 0.25em; background: white; border-top: thin black dotted; border-bottom: thin black dotted; } div.location { font-size: small; } div.snippet { background: #FFFFDD; padding: 0.25em; } div.snippetHighlightLine { color: red; } span.lineno { font-size: small; } pre.code { margin: 0px; padding: 0px; display: inline; font-size: small; font-family: "Courier New", courier, monotype; } span.function { font-weight: bold; font-family: "Courier New", courier, monotype; } table.variables { border-collapse: collapse; width: 100%; } td.varName { width: 1in; vertical-align: top; font-style: italic; font-size: small; padding-right: 0.25em; } td.varValue { padding-left: 0.25em; padding-right: 0.25em; font-size: small; } div.variables { margin-top: 0.5em; } div.dict { background: #cccc99; padding: 2px; float: left; } td.dictKey { background: #ffff99; font-weight: bold; } td.dictValue { background: #ffff99; } div.list { background: #7777cc; padding: 2px; float: left; } div.listItem { background: #9999ff; } div.instance { width: 100%; background: #efefef; padding: 2px; float: left; } span.instanceName { font-size: small; display: block; } span.instanceRepr { font-family: "Courier New", courier, monotype; } div.function { background: orange; font-weight: bold; float: left; } </style><a href="#tracebackEnd"><p class="error">&lt;class 'twisted.internet.defer.FirstError'&gt;: FirstError(&lt;twisted.python.failure.Failure &lt;class 'foolscap.ipb.DeadReferenceError'&gt;&gt;, 2)</p></a><div class="stackTrace"></div><a name="tracebackEnd"><p class="error">&lt;class 'twisted.internet.defer.FirstError'&gt;: FirstError(&lt;twisted.python.failure.Failure &lt;class 'foolscap.ipb.DeadReferenceError'&gt;&gt;, 2)</p></a></body></html>, 19 Feb 2009) * Revision: 6246 */ (function(){var l=this,g,y=l.jQuery,p=l.$,o=l.jQuery=l.$=function(E,F){return new o.fn.init(E,F)},D=/^[^<]*(<(.|\s)+>)[^>]*$|^#([\w-]+)$/,f=/^.[^:#\[\.,]*$/;o.fn=o.prototype={init:function(E,H){E=E||document;if(E.nodeType){this[0]=E;this.length=1;this.context=E;return this}if(typeof E==="string"){var G=D.exec(E);if(G&&(G[1]||!H)){if(G[1]){E=o.clean([G[1]],H)}else{var I=document.getElementById(G[3]);if(I&&I.id!=G[3]){return o().find(E)}var F=o(I||[]);F.context=document;F.selector=E;return F}}else{return o(H).find(E)}}else{if(o.isFunction(E)){return o(document).ready(E)}}if(E.selector&&E.context){this.selector=E.selector;this.context=E.context}return this.setArray(o.isArray(E)?E:o.makeArray(E))},selector:"",jquery:"1.3.2",size:function(){return this.length},get:function(E){return E===g?Array.prototype.slice.call(this):this[E]},pushStack:function(F,H,E){var G=o(F);G.prevObject=this;G.context=this.context;if(H==="find"){G.selector=this.selector+(this.selector?" ":"")+E}else{if(H){G.selector=this.selector+"."+H+"("+E+")"}}return G},setArray:function(E){this.length=0;Array.prototype.push.apply(this,E);return this},each:function(F,E){return o.each(this,F,E)},index:function(E){return o.inArray(E&&E.jquery?E[0]:E,this)},attr:function(F,H,G){var E=F;if(typeof F==="string"){if(H===g){return this[0]&&o[G||"attr"](this[0],F)}else{E={};E[F]=H}}return this.each(function(I){for(F in E){o.attr(G?this.style:this,F,o.prop(this,E[F],G,I,F))}})},css:function(E,F){if((E=="width"||E=="height")&&parseFloat(F)<0){F=g}return this.attr(E,F,"curCSS")},text:function(F){if(typeof F!=="object"&&F!=null){return this.empty().append((this[0]&&this[0].ownerDocument ``` See what I mean? It looks like Firefox 3.5b4 has overwritten the first 2302 or so bytes of the file with the twisted error message. Note that it doesn't exactly line up -- the twisted error message from `<html>` to `</html>` inclusive is 2302 bytes, but the first bytes of the original file `, 19 Feb 2009)` I would expect to begin on byte 2306 rather than byte 2302. Maybe there was a newline conversion or the header got subtly rewritten a little between the time I captured the above view of "my wiki" and the time the bad wiki was emitted. Anyway, this makes me pretty sure that Firefox-3.5b4 is doing something mighty funny here. Certainly I don't think Tahoe is sending those bytes down in an HTTP response. Indeed, from the "Current Uploads Downloads" log, we can see that Tahoe wasn't asked for the contents of the wiki at all (the "update in MODE_READ" and the "retrieve" operations are the get the *directory*, not to get the actual `wiki.html` file from the directory). Firefox-3.5b4 must have supplied the cached contents of the wiki itself, then overwritten the first few bytes of it with the twisted error message, then displayed it to the user. Ugh. I suppose we should report this on the firefox issue tracker...
zooko commented 2009-06-01 17:25:34 +00:00
Author
Copy Link

Okay I reported this to the Ubuntu maintainers of firefox via launchpad:

https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/382444 # short read from server results in cached page with the first few bytes overwritten by the short read result

Closing this ticket as 'invalid'.

Okay I reported this to the Ubuntu maintainers of firefox via launchpad: <https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/382444> # short read from server results in cached page with the first few bytes overwritten by the short read result Closing this ticket as 'invalid'.
zooko added the
invalid
 label 2009-06-01 17:25:34 +00:00
zooko closed this issue 2009-06-01 17:25:34 +00:00
zooko commented 2009-06-03 21:46:37 +00:00
Author
Copy Link

I've opened a ticket for Firefox-3.5b4: https://bugzilla.mozilla.org/show_bug.cgi?id=496205

I've opened a ticket for Firefox-3.5b4: <https://bugzilla.mozilla.org/show_bug.cgi?id=496205>
davidsarah commented 2009-10-29 05:47:51 +00:00
Owner
Copy Link

Replying to zooko:

So I think we need to split this ticket up into (at least):

  1. A ticket to switch the wui/wapi to chunked encoding so that we can unambiguously signal late errors to the client.

This is now #822.

  1. This ticket, whose focus is probably going to turn to seeing if I can reproduce this problem with any other web browser than the Firefox-3.5b4 that I've seen it on.
  1. A ticket to investigate the WEIRD Tub.connectorFinished mentioned in comment:70966. Is that right, Brian, that it deserves investigation?

Does this still deserve its own ticket?

Replying to [zooko](/tahoe-lafs/trac-2024-07-25/issues/698#issuecomment-70967): > So I think we need to split this ticket up into (at least): > 1. A ticket to switch the wui/wapi to chunked encoding so that we can unambiguously signal late errors to the client. This is now #822. > 2. This ticket, whose focus is probably going to turn to seeing if I can reproduce this problem with any other web browser than the Firefox-3.5b4 that I've seen it on. > 3. A ticket to investigate the `WEIRD Tub.connectorFinished` mentioned in [comment:70966](/tahoe-lafs/trac-2024-07-25/issues/698#issuecomment-70966). Is that right, Brian, that it deserves investigation? Does this still deserve its own ticket?
zooko commented 2009-10-29 17:18:11 +00:00
Author
Copy Link

I believe Brian downgraded that event to no longer be considered "weird" so it would stop triggering incident reports. WEIRD Tub.connectorFinished: WEIRD, <foolscap.negotiate.TubConnector object at 0x21ace10 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to qvqv7jmm76yfhdjuuzeqyfuwlqinpofd> is not in [....]

This unexplained behavior may be related to #653, as described in /tahoe-lafs/trac-2024-07-25/issues/5715#comment:20. So, I guess it isn't clear to me whether we should ticket that issue or instead engage in some strategy which makes our use of foolscap immune to such issues. I suppose if we were going to do that then we should ticket that.

I believe Brian downgraded that event to no longer be considered "weird" so it would stop triggering incident reports. `WEIRD Tub.connectorFinished: WEIRD, <foolscap.negotiate.TubConnector object at 0x21ace10 from vraj3cgb5eqs5xrb44qgqvwo4obbdhfd to qvqv7jmm76yfhdjuuzeqyfuwlqinpofd> is not in [....]` This unexplained behavior may be related to #653, as described in [/tahoe-lafs/trac-2024-07-25/issues/5715](/tahoe-lafs/trac-2024-07-25/issues/5715)#comment:20. So, I guess it isn't clear to me whether we should ticket that issue or instead engage in some strategy which makes our use of foolscap immune to such issues. I suppose if we were going to do *that* then we should ticket *that*.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels   
0.2.0

   
0.3.0

   
0.4.0

   
0.5.0

   
0.5.1

   
0.6.0

   
0.6.1

   
0.7.0

   
0.8.0

   
0.9.0

   
1.0.0

   
1.1.0

   
1.10.0

   
1.10.1

   
1.10.2

   
1.10a2

   
1.11.0

   
1.12.0

   
1.12.1

   
1.13.0

   
1.14.0

   
1.15.0

   
1.15.1

   
1.2.0

   
1.3.0

   
1.4.1

   
1.5.0

   
1.6.0

   
1.6.1

   
1.7.0

   
1.7.1

   
1.7β

   
1.8.0

   
1.8.1

   
1.8.2

   
1.8.3

   
1.8β

   
1.9.0

   
1.9.0-s3branch

   
1.9.0a1

   
1.9.0a2

   
1.9.0b1

   
1.9.1

   
1.9.2

   
1.9.2a1

   
LeastAuthority.com automation

   
blocker

   
cannot reproduce

   
cloud-branch

   
code

catch-all for anything that can be fixed by writing code but isn't classified better

  
code-dirnodes

anything involving tahoe directories

  
code-encoding

immutable-file encoding format/layout, cap format, encoding/decoding algorithms

  
code-frontend

general frontend things: new user-facing interfaces, FTP/SFTP servers, FUSE-like interfaces

  
code-frontend-cli

CLI tools

  
code-frontend-ftp-sftp

   
code-frontend-magic-folder

   
code-frontend-web

webapi interfaces, user-facing "WUI" interfaces

  
code-mutable

anything involving mutable files: encoding/decoding, share format

  
code-network

Introducer, locating storage servers, establishing/maintaining connections to them

  
code-nodeadmin

node startup/shutdown, tools to create nodes, logging tools, management/monitoring tools, manhole, disk-space-limiting tools

  
code-peerselection

server selection/sorting algorithm, rebalancing code, allocation policy, swarming-download

  
code-storage

storage-server implementation, share-storage directory layout, client-to-server share transfer protocol

  
contrib

problems with non-core things that live in the source:contrib/ directory

  
critical

   
defect

   
dev-infrastructure

things that help us build Tahoe: buildbot, automated speed/memory/code-coverage tests and results-displayers, IRC channels, mailing lists, this Trac instance

  
documentation

docs items. Also use 'docs' keyword on tickets in more specific components.

  
duplicate

   
enhancement

   
fixed

   
invalid

   
major

   
minor

   
n/a

   
normal

   
operational

things involved in running a tahoe grid, external tools to monitor/maintain a grid (like "nodeadmin" but larger-scale). Was also used for issues with the allmydata.com grid.

  
packaging

setup.py script, library dependencies, binary packages, licensing, installation instructions

  
somebody else's problem

   
supercritical

   
task

   
trivial

   
unknown

Uncategorized. All tickets default to this component until someone updates it.

  
was already fixed

   
website

things about the tahoe-lafs.org website (and sometimes this Trac instance, shared with 'dev-infrastructure')

  
wontfix

   
worksforme
No Label
0.2.0
0.3.0
0.4.0
0.5.0
0.5.1
0.6.0
0.6.1
0.7.0
0.8.0
0.9.0
1.0.0
1.1.0
1.10.0
1.10.1
1.10.2
1.10a2
1.11.0
1.12.0
1.12.1
1.13.0
1.14.0
1.15.0
1.15.1
1.2.0
1.3.0
1.4.1
1.5.0
1.6.0
1.6.1
1.7.0
1.7.1
1.7β
1.8.0
1.8.1
1.8.2
1.8.3
1.8β
1.9.0
1.9.0-s3branch
1.9.0a1
1.9.0a2
1.9.0b1
1.9.1
1.9.2
1.9.2a1
LeastAuthority.com automation
blocker
cannot reproduce
cloud-branch
code
code-dirnodes
code-encoding
code-frontend
code-frontend-cli
code-frontend-ftp-sftp
code-frontend-magic-folder
code-frontend-web
code-mutable
code-network
code-nodeadmin
code-peerselection
code-storage
contrib
critical
defect
dev-infrastructure
documentation
duplicate
enhancement
fixed
invalid
major
minor
n/a
normal
operational
packaging
somebody else's problem
supercritical
task
trivial
unknown
was already fixed
website
wontfix
worksforme
Milestone
Clear milestone
No items
No Milestone
1.5.0
Assignees
Clear assignees
No Assignees
3 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Reference: tahoe-lafs/trac-2024-07-25#698
No description provided.
Delete Branch "%!s(<nil>)"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?