Crashes on Treo 800w (WM6.1)

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Load prssr on Treo 800w.
2. Load multiple feeds in multiple categories via OPML input.
3. Attempt "Update All"

What is the expected output? What do you see instead?

Software crashes

What version of the product are you using? On what operating system?
prssr 1.4.2 on Windows Mobile 6.1 (Sprint Treo 800w)

Please provide any additional information below.

Original issue reported on code.google.com by kevin.se...@gmail.com on 16 Sep 2008 at 2:36

[GoogleCodeExporter]

Could you provide me with the OPML file, so that I'll be able to try this? It 
might
not be related just to Treo.

Original comment by and...@gmail.com on 16 Sep 2008 at 2:54

[GoogleCodeExporter]

"Software crashes"? Did you get error message(s) or was it a "silent exit" (see 
also
http://code.google.com/p/prssr/issues/detail?id=80&can=1)

Original comment by hfrmob...@gmail.com on 23 Sep 2008 at 8:06

[GoogleCodeExporter]

I have similar problem using 1.4.1 version, working with Samsung i780.  Exact 
same 
procecure and the phone will crash.

I did not see this problem using older 1.3.4

Original comment by leeman...@gmail.com on 9 Oct 2008 at 1:18

[GoogleCodeExporter]

Original comment by and...@gmail.com on 21 Oct 2008 at 4:30

• Added labels: Version-1.4.2

[GoogleCodeExporter]

I have an HTC Touch, with Win Mobile 6, in Portuguese and prss crashes as well,
silently. Do you want the opml file? I sense it ha something to do with 
international
feeds, with accents...

Original comment by pedro.md...@gmail.com on 25 Oct 2008 at 9:25

[GoogleCodeExporter]

@pedro.mdl.santos (see also 
http://code.google.com/p/prssr/issues/detail?id=80&can=1)

I previously used the HTC Touch (WM6) (German) and also suffered from "silent 
exits".
Did the device run low on memory when silent exits occurred?

Original comment by hfrmob...@gmail.com on 25 Oct 2008 at 10:32

[GoogleCodeExporter]

pedro.mdl.santos: Yes my comment #1 is still valid. Still waiting for the opml 
file
to try this out. Without it, I can not do much.

Original comment by and...@gmail.com on 25 Oct 2008 at 4:31

[GoogleCodeExporter]

Attached is my OMPL file.

Original comment by leeman...@gmail.com on 30 Oct 2008 at 4:41

Attachments:

• myChannels.opml

[GoogleCodeExporter]

So I tried this (with the latest development version). I did a fresh install of
pRSSreader, started Site manager, Menu | OPML file | Import, then add the files 
(left
the append option checked), ok, Update All.

I saw 8 channels, same as in the posted OPML file. Everything went ok. I did 
not see
any news items for second and the fourth channel. But no crashes.

Original comment by and...@gmail.com on 8 Nov 2008 at 9:18

[GoogleCodeExporter]

I'd be willing to give it a go with a developer version, if it will help to 
capture 
some information.

Original comment by bogg...@gmail.com on 13 Nov 2008 at 4:30

[GoogleCodeExporter]

1. Launch pRSSreader
2. Select Update All
3. pRSSreader exits without warning

prsslog.txt file is attached

I am not syncing with another reader/service, only the rss feeds in my attached 
opml 
file.

Original comment by bogg...@gmail.com on 13 Nov 2008 at 4:56

Attachments:

• Feeds.opml
• prsslog.txt

[GoogleCodeExporter]

I have tried and been successful in reproducing this behavior with:
1.4.2
b138
1.4.3

If I drop back to 1.4.1, the behavior vanishes.  All other variables are 
identical 
(ROM setup, OPML import, etc.)

Best regards,
-boggsie

Original comment by bogg...@gmail.com on 13 Nov 2008 at 6:20

[GoogleCodeExporter]

dump files from crash 

Original comment by bogg...@gmail.com on 14 Nov 2008 at 1:56

Attachments:

• cedevice.log
• Ce111308-01.kdmp

[GoogleCodeExporter]

boogsie: I tried your feed list, but it worked for me. I looked in the log file 
you
sent, but there is not much info. Could you switch to the LogLevel 5, replicate 
the
crash and submit the log file here again - compress it if it is too large (ZIP 
works
for me).

What I see in the log file is that the last function is CFeedSync::MergeFeed(). 
What
are your settings for the feed caching/archive (both the global and for feeds)?

Did you try to get rid of pRSSreader (completely) and do a fresh install and a 
fresh
import of your feeds?

Original comment by and...@gmail.com on 17 Nov 2008 at 1:22

[GoogleCodeExporter]

Thank you for your reply.

In the enclosed archive (*.zip) I have a couple of log files along with images 
of 
each of my settings screens.

In prior research / testing I was careful to re-flash my base rom image and 
then 
reinstall pRSSreader ... I wanted to do what I could to ensure a clean 
environment 
for evaluating the behavior I have been experiencing.

Original comment by bogg...@gmail.com on 17 Nov 2008 at 1:49

Attachments:

• pRSSreader.zip

[GoogleCodeExporter]

andrsd: What device and version of ce/os are you using?  I ask because this 
might be 
a 6.1 issue.

Original comment by bogg...@gmail.com on 17 Nov 2008 at 1:51

[GoogleCodeExporter]

Sure it can. My device is Loox 718, WM2003SE. I'm wondering what is the problem,
since I'm not able to reproduce this in any way. It is also ok in the emulator 
(WM6).
I recently downloaded images for WM6.1, so I'll also try these.

What I saw some times was, that users used a custom ROM image on their device, 
but
when they got back to the original, the problem went away... I'm not able to do 
much
 with such cases, since I do not have the problematic device on my desk to debug,
etc. That was one reason why I opened the source code, to allow others to 
compile the
source and try to track down the problem on their device.

Original comment by and...@gmail.com on 17 Nov 2008 at 3:19

[GoogleCodeExporter]

Originally, I suspected a custom rom and the last thing I wanted to do is send 
you 
on a wild goose chase to fix a problem with an XIP or SYS customization.  So, 
to 
alleviate this issue, I flashed back to a standard HTC rom (Vogue 3.14) 
5.2.19212.

What is so interesting (irritating) about this, is that with the same ROM 
version 
(referenced above) and setup (1.4.3), I failed to successfully get updates all 
day 
on Saturday and most of the day on Sunday.  Then, at some point late Sunday, 
things 
started to work and I was able to receive updates.  Occasionally the program 
would 
close while reading feeds, but sending and recieving would not immediately fail 
as I 
had reported.

Earlier today, the program successfully retrieved >350 feeds and I have been 
updating (retrieving feeds) and reading throughout the morning ... no changes, 
no 
reset (soft or otherwise) and I haven't experienced any issues.

<shrugs>

Thanks for the attention to the issue, though.

Best regards,
-boggsie

Original comment by bogg...@gmail.com on 17 Nov 2008 at 3:36

[GoogleCodeExporter]

<banging head on desk>

Well, I've had updates set to automatic on 90-minute intervals.  So far, I have 
received rss updates on three occasions and I haven't had or experienced any 
issues, 
whatsoever.

</banging head on desk>

I'm happy that all is working well, but I know something is there ... whether 
it is 
a WM6.1 issue, a data / connectivity issue, a feed issue or a pRSSreader issue.

My experience over the last 24 hours tends to preclude WM6.1 and pRSSreader as 
the 
culprit and if you are able to retrieve the feeds in my opml, then it probably 
isn't 
a feed issue.  I wonder what kind of instability might be introduced if there 
are 
data or connectivity issues?

Best regards,
-boggsie

Original comment by bogg...@gmail.com on 17 Nov 2008 at 8:41

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Hi!

I see the same on Loox 720 (WinCE 2003SE) with original ROM. It happens on 
UpdateAll
and Cache Unread when caching the enclosure of particular feeds. I attached my 
opml
file and the log with level 5 on prssr 1.4.3.
The download completes. Then I see the status bar is refreshing several times 
until
the crash happens silently.

Original comment by Marten.B...@gmx.de on 7 Dec 2008 at 7:51

Attachments:

• myChannels.opml
• prssr.log

[GoogleCodeExporter]

I found special condition with this feed: 
I had both per-feed options enabled: 
"Cache articles for offline reading" and 
"Cache enclosures". The crash happens and 
I see "Downloading 1/2 .." in the status 
bar. The first downloaded file is also 
the enclosure! Using this configuration 
with other feeds I see that the first 
downloaded file is the article.

As a workaround, I disabled caching of 
the article, but keep caching the 
enclosure. No crashes in that case.

Original comment by Marten.B...@gmx.de on 12 Dec 2008 at 10:58

[GoogleCodeExporter]

simply amazing, Marten ...

I had all but given up using pRSSreader ... I was working on getting my 
development 
environment up and running to try to help with this issue.

I read Marten's theory, fired up pRSSreader 1.4.3 and can confirm ...
with both cache options set, pRSSreader (silent) crash 4 times in a row
with only cache articles set, pRSSreader (silent) crash 3 times in a row
with only cache images within the feed set, pRSSreader did not crash

Again, excellent job of nailing this behavior, Marten!!

Best regards,
-boggsie

Original comment by bogg...@gmail.com on 13 Dec 2008 at 3:25

[GoogleCodeExporter]

Marten: The rss feed of 1liveklubbing is linking the audio file in two different
ways: first in the <link> tag, which is usually used for the HTML file with the
article and then in <enclosure> tag, which is used for enclosures.

pRSSreader expects that files specified by <link> tag are HTML files and is 
doing
some transformations of the HTML file for offline viewing. That might be the 
reason
why it fails for you. I tried to replicate this, but the way thought <link> tag 
did
not come through, pRSSreader was not able to download the file and reported an 
error.
But it was able to download the enclosure.

Original comment by and...@gmail.com on 13 Dec 2008 at 3:48

[GoogleCodeExporter]

I had the HTML optimizer turned on, so I had different environment than you. 
Turning
it off, I replicated the problem. It really is in that pRSSreader treats the 
first
file as HTML file, instead of the HTML file.

Original comment by and...@gmail.com on 13 Dec 2008 at 3:58

[GoogleCodeExporter]

So I fixed the Marten's problem (r207)

Original comment by and...@gmail.com on 13 Dec 2008 at 6:02