Community Discussions and Support
Pegasus crashes when replying to emails (from Outlook users?)

We use Pegasus 4.80 over a network, several users, nothing that complicated IMHO, although we have been using Pegasus a long time now... Server and clients are mostly Win 10 Pro.


Since updating to 4.80, there have been a very few emails where, if you click on Reply or Forward, Pegasus instantly crashes. Bounces however are no problem.


I've tried the same from several user's accounts and devices, and get the same result, so it's not likely to be an OS issue.


Is it attachments? Some of the emails have attachments, but not all.


Is it the addressee? No, we can email them ordinarily, and even reply to other emails, so it appears to be something that only occurs when the specific odd email is replied to.


In the headers of the emails, I can see some common factors that indicate the sender uses Office365/Outlook, perhaps in conjunction with Exchange. No idea if these are relevant or interesting, but I'll share some if someone thinks it might help.


Finally, the crash usually produces a couple of entries in Event Viewer:


=======
Log Name: Application
Source: Application Error
Date: 01/09/2022 15:07:57
Event ID: 1000
Task Category: (100)
Level: Error
Keywords: Classic
User: N/A
Computer: SUPPORT14
Description:
Faulting application name: winpm-32.exe, version: 4.8.0.1028, time stamp: 0x61e3dbc2
Faulting module name: winpm-32.exe, version: 4.8.0.1028, time stamp: 0x61e3dbc2
Exception code: 0xc0000409
Fault offset: 0x001c2613
Faulting process ID: 0x2720
Faulting application start time: 0x01d8bd0ec0dac55d
Faulting application path: \Fs2019\Pegasus\Programs\winpm-32.exe
Faulting module path: \Fs2019\Pegasus\Programs\winpm-32.exe
Report ID: 3b33a0e8-c78a-4526-9924-98b0e80b1a25
Faulting package full name:
Faulting package-relative application ID:
Event Xml:


<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Application Error" />
<EventID Qualifiers="0">1000</EventID>
<Version>0</Version>
<Level>2</Level>
<Task>100</Task>
<Opcode>0</Opcode>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2022-09-01T14:07:57.0771079Z" />
<EventRecordID>176559</EventRecordID>
<Correlation />
<Execution ProcessID="0" ThreadID="0" />
<Channel>Application</Channel>
<Computer>SUPPORT14</Computer>
<Security />
</System>
<EventData>
<Data>winpm-32.exe</Data>
<Data>4.8.0.1028</Data>
<Data>61e3dbc2</Data>
<Data>winpm-32.exe</Data>
<Data>4.8.0.1028</Data>
<Data>61e3dbc2</Data>
<Data>c0000409</Data>
<Data>001c2613</Data>
<Data>2720</Data>
<Data>01d8bd0ec0dac55d</Data>
<Data>\Fs2019\Pegasus\Programs\winpm-32.exe</Data>
<Data>\Fs2019\Pegasus\Programs\winpm-32.exe</Data>
<Data>3b33a0e8-c78a-4526-9924-98b0e80b1a25</Data>
<Data>
</Data>
<Data>
</Data>
</EventData>
</Event>


===========
Log Name: Application
Source: Windows Error Reporting
Date: 01/09/2022 15:08:06
Event ID: 1001
Task Category: None
Level: Information
Keywords: Classic
User: N/A
Computer: SUPPORT14
Description:
Fault bucket 1327807070259735361, type 5
Event Name: BEX
Response: Not available
Cab Id: 0


Problem signature:
P1: winpm-32.exe
P2: 4.8.0.1028
P3: 61e3dbc2
P4: winpm-32.exe
P5: 4.8.0.1028
P6: 61e3dbc2
P7: 001c2613
P8: c0000409
P9: 00000000
P10:


Attached files:
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER3C20.tmp.dmp
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4EAF.tmp.WERInternalMetadata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FBA.tmp.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FC8.tmp.csv
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER5036.tmp.txt


These files may be available here:
\?\C:\ProgramData\Microsoft\Windows\WER\ReportArchive\AppCrash_winpm-32.exe_ed53637fe059866abffbf929e62235dbd83b1358_bc21b527_cad656ff-56ba-4e7a-88cc-e82c5cf29495


Analysis symbol:
Rechecking for solution: 0
Report Id: 3b33a0e8-c78a-4526-9924-98b0e80b1a25
Report Status: 268435456
Hashed bucket: c422e79174d111dbd26d517ed1156b41
Cab Guid: 0
Event Xml:


<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Windows Error Reporting" />
<EventID Qualifiers="0">1001</EventID>
<Version>0</Version>
<Level>4</Level>
<Task>0</Task>
<Opcode>0</Opcode>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2022-09-01T14:08:06.4418434Z" />
<EventRecordID>176560</EventRecordID>
<Correlation />
<Execution ProcessID="0" ThreadID="0" />
<Channel>Application</Channel>
<Computer>SUPPORT14</Computer>
<Security />
</System>
<EventData>
<Data>1327807070259735361</Data>
<Data>5</Data>
<Data>BEX</Data>
<Data>Not available</Data>
<Data>0</Data>
<Data>winpm-32.exe</Data>
<Data>4.8.0.1028</Data>
<Data>61e3dbc2</Data>
<Data>winpm-32.exe</Data>
<Data>4.8.0.1028</Data>
<Data>61e3dbc2</Data>
<Data>001c2613</Data>
<Data>c0000409</Data>
<Data>00000000</Data>
<Data>
</Data>
<Data>
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER3C20.tmp.dmp
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4EAF.tmp.WERInternalMetadata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FBA.tmp.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FC8.tmp.csv
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER5036.tmp.txt</Data>
<Data>\?\C:\ProgramData\Microsoft\Windows\WER\ReportArchive\AppCrash_winpm-32.exe_ed53637fe059866abffbf929e62235dbd83b1358_bc21b527_cad656ff-56ba-4e7a-88cc-e82c5cf29495</Data>
<Data>
</Data>
<Data>0</Data>
<Data>3b33a0e8-c78a-4526-9924-98b0e80b1a25</Data>
<Data>268435456</Data>
<Data>c422e79174d111dbd26d517ed1156b41</Data>
<Data>0</Data>
</EventData>
</Event>


===========


Any thoughts/ideas?


We use Pegasus 4.80 over a network, several users, nothing that complicated IMHO, although we have been using Pegasus a long time now... Server and clients are mostly Win 10 Pro. Since updating to 4.80, there have been a very few emails where, if you click on Reply or Forward, Pegasus instantly crashes. Bounces however are no problem. I&#039;ve tried the same from several user&#039;s accounts and devices, and get the same result, so it&#039;s not likely to be an OS issue. Is it attachments? Some of the emails have attachments, but not all. Is it the addressee? No, we can email them ordinarily, and even reply to other emails, so it appears to be something that only occurs when the specific odd email is replied to. In the headers of the emails, I can see some common factors that indicate the sender uses Office365/Outlook, perhaps in conjunction with Exchange. No idea if these are relevant or interesting, but I&#039;ll share some if someone thinks it might help. Finally, the crash usually produces a couple of entries in Event Viewer: ======= Log Name: Application Source: Application Error Date: 01/09/2022 15:07:57 Event ID: 1000 Task Category: (100) Level: Error Keywords: Classic User: N/A Computer: SUPPORT14 Description: Faulting application name: winpm-32.exe, version: 4.8.0.1028, time stamp: 0x61e3dbc2 Faulting module name: winpm-32.exe, version: 4.8.0.1028, time stamp: 0x61e3dbc2 Exception code: 0xc0000409 Fault offset: 0x001c2613 Faulting process ID: 0x2720 Faulting application start time: 0x01d8bd0ec0dac55d Faulting application path: \\Fs2019\Pegasus\Programs\winpm-32.exe Faulting module path: \\Fs2019\Pegasus\Programs\winpm-32.exe Report ID: 3b33a0e8-c78a-4526-9924-98b0e80b1a25 Faulting package full name: Faulting package-relative application ID: Event Xml: &lt;Event xmlns=&quot;http://schemas.microsoft.com/win/2004/08/events/event&quot;&gt; &lt;System&gt; &lt;Provider Name=&quot;Application Error&quot; /&gt; &lt;EventID Qualifiers=&quot;0&quot;&gt;1000&lt;/EventID&gt; &lt;Version&gt;0&lt;/Version&gt; &lt;Level&gt;2&lt;/Level&gt; &lt;Task&gt;100&lt;/Task&gt; &lt;Opcode&gt;0&lt;/Opcode&gt; &lt;Keywords&gt;0x80000000000000&lt;/Keywords&gt; &lt;TimeCreated SystemTime=&quot;2022-09-01T14:07:57.0771079Z&quot; /&gt; &lt;EventRecordID&gt;176559&lt;/EventRecordID&gt; &lt;Correlation /&gt; &lt;Execution ProcessID=&quot;0&quot; ThreadID=&quot;0&quot; /&gt; &lt;Channel&gt;Application&lt;/Channel&gt; &lt;Computer&gt;SUPPORT14&lt;/Computer&gt; &lt;Security /&gt; &lt;/System&gt; &lt;EventData&gt; &lt;Data&gt;winpm-32.exe&lt;/Data&gt; &lt;Data&gt;4.8.0.1028&lt;/Data&gt; &lt;Data&gt;61e3dbc2&lt;/Data&gt; &lt;Data&gt;winpm-32.exe&lt;/Data&gt; &lt;Data&gt;4.8.0.1028&lt;/Data&gt; &lt;Data&gt;61e3dbc2&lt;/Data&gt; &lt;Data&gt;c0000409&lt;/Data&gt; &lt;Data&gt;001c2613&lt;/Data&gt; &lt;Data&gt;2720&lt;/Data&gt; &lt;Data&gt;01d8bd0ec0dac55d&lt;/Data&gt; &lt;Data&gt;\\Fs2019\Pegasus\Programs\winpm-32.exe&lt;/Data&gt; &lt;Data&gt;\\Fs2019\Pegasus\Programs\winpm-32.exe&lt;/Data&gt; &lt;Data&gt;3b33a0e8-c78a-4526-9924-98b0e80b1a25&lt;/Data&gt; &lt;Data&gt; &lt;/Data&gt; &lt;Data&gt; &lt;/Data&gt; &lt;/EventData&gt; &lt;/Event&gt; =========== Log Name: Application Source: Windows Error Reporting Date: 01/09/2022 15:08:06 Event ID: 1001 Task Category: None Level: Information Keywords: Classic User: N/A Computer: SUPPORT14 Description: Fault bucket 1327807070259735361, type 5 Event Name: BEX Response: Not available Cab Id: 0 Problem signature: P1: winpm-32.exe P2: 4.8.0.1028 P3: 61e3dbc2 P4: winpm-32.exe P5: 4.8.0.1028 P6: 61e3dbc2 P7: 001c2613 P8: c0000409 P9: 00000000 P10: Attached files: \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER3C20.tmp.dmp \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4EAF.tmp.WERInternalMetadata.xml \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FBA.tmp.xml \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FC8.tmp.csv \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER5036.tmp.txt These files may be available here: \\?\C:\ProgramData\Microsoft\Windows\WER\ReportArchive\AppCrash_winpm-32.exe_ed53637fe059866abffbf929e62235dbd83b1358_bc21b527_cad656ff-56ba-4e7a-88cc-e82c5cf29495 Analysis symbol: Rechecking for solution: 0 Report Id: 3b33a0e8-c78a-4526-9924-98b0e80b1a25 Report Status: 268435456 Hashed bucket: c422e79174d111dbd26d517ed1156b41 Cab Guid: 0 Event Xml: &lt;Event xmlns=&quot;http://schemas.microsoft.com/win/2004/08/events/event&quot;&gt; &lt;System&gt; &lt;Provider Name=&quot;Windows Error Reporting&quot; /&gt; &lt;EventID Qualifiers=&quot;0&quot;&gt;1001&lt;/EventID&gt; &lt;Version&gt;0&lt;/Version&gt; &lt;Level&gt;4&lt;/Level&gt; &lt;Task&gt;0&lt;/Task&gt; &lt;Opcode&gt;0&lt;/Opcode&gt; &lt;Keywords&gt;0x80000000000000&lt;/Keywords&gt; &lt;TimeCreated SystemTime=&quot;2022-09-01T14:08:06.4418434Z&quot; /&gt; &lt;EventRecordID&gt;176560&lt;/EventRecordID&gt; &lt;Correlation /&gt; &lt;Execution ProcessID=&quot;0&quot; ThreadID=&quot;0&quot; /&gt; &lt;Channel&gt;Application&lt;/Channel&gt; &lt;Computer&gt;SUPPORT14&lt;/Computer&gt; &lt;Security /&gt; &lt;/System&gt; &lt;EventData&gt; &lt;Data&gt;1327807070259735361&lt;/Data&gt; &lt;Data&gt;5&lt;/Data&gt; &lt;Data&gt;BEX&lt;/Data&gt; &lt;Data&gt;Not available&lt;/Data&gt; &lt;Data&gt;0&lt;/Data&gt; &lt;Data&gt;winpm-32.exe&lt;/Data&gt; &lt;Data&gt;4.8.0.1028&lt;/Data&gt; &lt;Data&gt;61e3dbc2&lt;/Data&gt; &lt;Data&gt;winpm-32.exe&lt;/Data&gt; &lt;Data&gt;4.8.0.1028&lt;/Data&gt; &lt;Data&gt;61e3dbc2&lt;/Data&gt; &lt;Data&gt;001c2613&lt;/Data&gt; &lt;Data&gt;c0000409&lt;/Data&gt; &lt;Data&gt;00000000&lt;/Data&gt; &lt;Data&gt; &lt;/Data&gt; &lt;Data&gt; \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER3C20.tmp.dmp \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4EAF.tmp.WERInternalMetadata.xml \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FBA.tmp.xml \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER4FC8.tmp.csv \\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER5036.tmp.txt&lt;/Data&gt; &lt;Data&gt;\\?\C:\ProgramData\Microsoft\Windows\WER\ReportArchive\AppCrash_winpm-32.exe_ed53637fe059866abffbf929e62235dbd83b1358_bc21b527_cad656ff-56ba-4e7a-88cc-e82c5cf29495&lt;/Data&gt; &lt;Data&gt; &lt;/Data&gt; &lt;Data&gt;0&lt;/Data&gt; &lt;Data&gt;3b33a0e8-c78a-4526-9924-98b0e80b1a25&lt;/Data&gt; &lt;Data&gt;268435456&lt;/Data&gt; &lt;Data&gt;c422e79174d111dbd26d517ed1156b41&lt;/Data&gt; &lt;Data&gt;0&lt;/Data&gt; &lt;/EventData&gt; &lt;/Event&gt; =========== Any thoughts/ideas?
edited Sep 1 '22 at 3:47 pm

Exception code: 0xc0000409


Sorry, ferret without a name, most of this information (except for the line above) is only relevant to MS and its development tools which the most likely failing part of Pegasus Mail isn't written by or with.


Anyway, first thing to do after updating Pegasus Mail to v4.8 would be an update of its IERenderer module which you should have been notified of by IER's toolbar button topright on Pegasus Mail's main toolbar. If this has already been applied and doesn't help I would be glad to get a sample message for looking into this directed towards < beta-reports [at] pmail.gen.nz > easiest done by forwarding as an attachment (according to the forward options dialog). This way it shouldn't crash nor get modified by editing. If doing so please add a meaningful subject for the message you send (like a reference to this thread or even the same subject) as this account is flooded with spam and I need to easily tell it apart from this for not deleting any.


If forwarding isn't an option I'll make another suggestion, but trying to duplicate your issue in my development environment is the fastest way for fixing.


[quote=&quot;pid:54408, uid:28403&quot;]Exception code: 0xc0000409[/quote] Sorry, ferret without a name, most of this information (except for the line above) is only relevant to MS and its development tools which the most likely failing part of Pegasus Mail isn&#039;t written by or with. Anyway, first thing to do after updating Pegasus Mail to v4.8 would be an update of its IERenderer module which you should have been notified of by IER&#039;s toolbar button topright on Pegasus Mail&#039;s main toolbar. If this has already been applied and doesn&#039;t help I would be glad to get a sample message for looking into this directed towards &lt; beta-reports [at] pmail.gen.nz &gt; easiest done by forwarding as an attachment (according to the forward options dialog). This way it shouldn&#039;t crash nor get modified by editing. If doing so please add a meaningful subject for the message you send (like a reference to this thread or even the same subject) as this account is flooded with spam and I need to easily tell it apart from this for not deleting any. If forwarding isn&#039;t an option I&#039;ll make another suggestion, but trying to duplicate your issue in my development environment is the fastest way for fixing.
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C

Would agree to make sure you have the IERender latest version updated first, and see if that resolves issue..


If not resolved.


  1. Do the machines have C:\TEMP directory and do the users have read/write access. Bearhtml uses that directory, and some of the newer setups don't have it, or have restricted access to it.

If there is an email you could share would be interesting to see if there is something strange in it.
I run Pegasus under linux using wine, and don't have IERender or BearHTML active. So, just have the TER/HTS running..


You might want to check the users temp directory, and see the latest files in it that might show something.


Might also be an antivirus program that is grabbing files during the creation of reply options?


Would agree to make sure you have the IERender latest version updated first, and see if that resolves issue.. If not resolved. 1. Do the machines have C:\TEMP directory and do the users have read/write access. Bearhtml uses that directory, and some of the newer setups don&#039;t have it, or have restricted access to it. If there is an email you could share would be interesting to see if there is something strange in it. I run Pegasus under linux using wine, and don&#039;t have IERender or BearHTML active. So, just have the TER/HTS running.. You might want to check the users temp directory, and see the latest files in it that might show something. Might also be an antivirus program that is grabbing files during the creation of reply options?

mikes@guam.net

Thanks guys.


OK, updating IERenderer reduced the number of failing emails by half! I've forwarded a sample of the still-failing email on, as requested.


In answer to the other questions,


  1. I have full R/W to C:\temp, and there's nothing recent, unusual or unexpected in there.
  2. I use Windows Defender on Win10Pro. Other users use different AV products (AVG and Avast). We haven't seen any alerts when sending emails.

Thanks guys. OK, updating IERenderer reduced the number of failing emails by half! I&#039;ve forwarded a sample of the still-failing email on, as requested. In answer to the other questions, 1. I have full R/W to C:\temp, and there&#039;s nothing recent, unusual or unexpected in there. 2. I use Windows Defender on Win10Pro. Other users use different AV products (AVG and Avast). We haven&#039;t seen any alerts when sending emails.

Thanks for reply. Usually, the problem with the antivirus stuff is that they grab control of the files, and don't allow Pegasus to access the files like it expects to be able too. Thus causing failures. At least that is how I've understood it.


Don't know how the best process would be. Seems to have an option to put in link or upload file.
Generally, I would just have person forward message unedited to my mikes@guam.net email.


But would be interesting to see what might be inside message that causes issue.


Thanks and Good Luck.


Thanks for reply. Usually, the problem with the antivirus stuff is that they grab control of the files, and don&#039;t allow Pegasus to access the files like it expects to be able too. Thus causing failures. At least that is how I&#039;ve understood it. Don&#039;t know how the best process would be. Seems to have an option to put in link or upload file. Generally, I would just have person forward message unedited to my mikes@guam.net email. But would be interesting to see what might be inside message that causes issue. Thanks and Good Luck.

mikes@guam.net

Thanks for reply. Usually, the problem with the antivirus stuff is that they grab control of the files, and don't allow Pegasus to access the files like it expects to be able too. Thus causing failures. At least that is how I've understood it.


That is exactly it. The advice is to exclude the mailbox directories from active scanning (what Windows Defender calls "Real-time trotection" ) to insure Windows Defender, or any AV product, does not attempt to grab a file for scanning while Pegasus Mail is in the process of updating it. If the Pegasus Mail installation is to the default location of C:\PMAIL with mailboxes in C:\PMAIL\MAIL then excluding C:\PMAIL is easiest. That is what I have done in my Windows Defender.


Regarding failing replies, in May a problem was identified that caused crashes on replies to messages that contained extremely long lines. The crashes occurred only when there was no long-line handling in the reply. Enabling either Wrap or Reformat prevented the crash. I don't know if this has been addressed in a update to IERenderer. I recommend testing some of the failure messages paying specific attention to long-line handling.


[quote=&quot;pid:54416, uid:2546&quot;]Thanks for reply. Usually, the problem with the antivirus stuff is that they grab control of the files, and don&#039;t allow Pegasus to access the files like it expects to be able too. Thus causing failures. At least that is how I&#039;ve understood it.[/quote] That is exactly it. The advice is to exclude the mailbox directories from active scanning (what Windows Defender calls &quot;Real-time trotection&quot; ) to insure Windows Defender, or any AV product, does not attempt to grab a file for scanning while Pegasus Mail is in the process of updating it. If the Pegasus Mail installation is to the default location of C:\PMAIL with mailboxes in C:\PMAIL\MAIL then excluding C:\PMAIL is easiest. That is what I have done in my Windows Defender. Regarding failing replies, in May a problem was identified that caused crashes on replies to messages that contained extremely long lines. The crashes occurred only when there was no long-line handling in the reply. Enabling either Wrap or Reformat prevented the crash. I don&#039;t know if this has been addressed in a update to IERenderer. I recommend testing some of the failure messages paying specific attention to long-line handling.

I don't know if this has been addressed in a update to IERenderer. I recommend testing some of the failure messages paying specific attention to long-line handling.

If there was something I could do I certainly should have done so, but there's more involved than just the HTML renderer since the MIME decoding unfolding those long lines is done by Pegasus Mail proper before handing them over to any HTML renderer which might break long lines in such a way that this cannot by fixed anymore unless David Harris himself takes care of it by enlarging his buffer sizes.


[quote=&quot;pid:54418, uid:28772&quot;]I don&#039;t know if this has been addressed in a update to IERenderer. I recommend testing some of the failure messages paying specific attention to long-line handling.[/quote] If there was something I could do I certainly should have done so, but there&#039;s more involved than just the HTML renderer since the MIME decoding unfolding those long lines is done by Pegasus Mail proper _before_ handing them over to any HTML renderer which might break long lines in such a way that this cannot by fixed anymore unless David Harris himself takes care of it by enlarging his buffer sizes.
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C

OK, updating IERenderer reduced the number of failing emails by half! I've forwarded a sample of the still-failing email on, as requested.


Ok, got it, its a really weird one, no HTML involved: I can write a reply and send it to queue but the first weirdness is the editor doesn't close as it is supposed to. The draft goes to the queue but when trying to deliver it via SMTP Pegasus Mail finally crashes on my Win7 machine. Definitely a case for David Harris himself, I'll forward it to him. Thanks for providing it!


[quote=&quot;pid:54412, uid:28403&quot;]OK, updating IERenderer reduced the number of failing emails by half! I&#039;ve forwarded a sample of the still-failing email on, as requested.[/quote] Ok, got it, its a really weird one, no HTML involved: I can write a reply and send it to queue but the first weirdness is the editor doesn&#039;t close as it is supposed to. The draft goes to the queue but when trying to deliver it via SMTP Pegasus Mail finally crashes on my Win7 machine. Definitely a case for David Harris himself, I&#039;ll forward it to him. Thanks for providing it!
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C

Would be interesting to see the email?
Know that many many years ago, had an uucp gateway for Pegasus mail DOS version on the Novell 2.2 networks. Had it on site, and then got messages that people had issues with some emails. Turned out the Outlook was sending out emails with lines longer than the RFC limit of 1022 I believe. 1022+cr+lf to make 1024. So was causing a buffer overflow. No way to get Microsoft/Outlook to follow the RFC, so just increased the buffer size to handle the long lines. Microsoft decided to write whole paragraphs as a single long line...


Had some wierd messages that had URLs that were longer than 1024 characters that caused issues, but not much can be done for that.


Hopefully, a full solution can be found.


Would be interesting to see the email? Know that many many years ago, had an uucp gateway for Pegasus mail DOS version on the Novell 2.2 networks. Had it on site, and then got messages that people had issues with some emails. Turned out the Outlook was sending out emails with lines longer than the RFC limit of 1022 I believe. 1022+cr+lf to make 1024. So was causing a buffer overflow. No way to get Microsoft/Outlook to follow the RFC, so just increased the buffer size to handle the long lines. Microsoft decided to write whole paragraphs as a single long line... Had some wierd messages that had URLs that were longer than 1024 characters that caused issues, but not much can be done for that. Hopefully, a full solution can be found.

mikes@guam.net

Regarding failing replies, in May a problem was identified that caused crashes on replies to messages that contained extremely long lines. The crashes occurred only when there was no long-line handling in the reply. Enabling either Wrap or Reformat prevented the crash. I don't know if this has been addressed in a update to IERenderer. I recommend testing some of the failure messages paying specific attention to long-line handling.


I just tested the remaining failing mails, and they all worked correctly using Wrap and Reformat, but not None. Thanks for that! And yes, the emails did contain some very long lines...


[quote=&quot;pid:54418, uid:28772&quot;]Regarding failing replies, in May a problem was identified that caused crashes on replies to messages that contained extremely long lines. The crashes occurred only when there was no long-line handling in the reply. Enabling either Wrap or Reformat prevented the crash. I don&#039;t know if this has been addressed in a update to IERenderer. I recommend testing some of the failure messages paying specific attention to long-line handling.[/quote] I just tested the remaining failing mails, and they all worked correctly using Wrap and Reformat, but not None. Thanks for that! And yes, the emails did contain some very long lines...
edited Sep 6 '22 at 9:29 am

I just tested the remaining failing mails, and they all worked correctly using Wrap and Reformat, but not None. Thanks for that! And yes, the emails did contain some very long lines...

Yep, the quoted-printable message transport encoding hides this by applying a proper folding for transport but by unfolding it for including into replies it creates the whole message becoming one long line of 1842 characters. I'm really surprised that this issue (regardless of being HTML involved or not) is still around after so many years of Pegasus Mail's history ... Anyway, I'm going to send a reminder to David Harris after removing the attachment.


[quote=&quot;pid:54424, uid:28403&quot;]I just tested the remaining failing mails, and they all worked correctly using Wrap and Reformat, but not None. Thanks for that! And yes, the emails did contain some very long lines...[/quote] Yep, the quoted-printable message transport encoding hides this by applying a proper folding for transport but by unfolding it for including into replies it creates the whole message becoming one long line of 1842 characters. I&#039;m really surprised that this issue (regardless of being HTML involved or not) is still around after so many years of Pegasus Mail&#039;s history ... Anyway, I&#039;m going to send a reminder to David Harris after removing the attachment.
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C
edited Sep 6 '22 at 11:16 am

I'm really surprised that this issue (regardless of being HTML involved or not) is still around after so many years of Pegasus Mail's history ...

From David's reply:



Oddly enough, this particular problem would have been there for many years, but
would only have occurred in the situation where the line was illegally long when
QP-decoded and not wrapped in the editor, indicating that most people probably
aren't encountering it.



IOW: He's working on a fix (appearing to be not as easy as it sounds).


[quote=&quot;pid:54425, uid:2133&quot;]I&#039;m really surprised that this issue (regardless of being HTML involved or not) is still around after so many years of Pegasus Mail&#039;s history ...[/quote] From David&#039;s reply: &gt; Oddly enough, this particular problem would have been there for many years, but would only have occurred in the situation where the line was illegally long when QP-decoded and not wrapped in the editor, indicating that most people probably aren&#039;t encountering it. IOW: He&#039;s working on a fix (appearing to be not as easy as it sounds).
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C

Hi,


Still today it seems this problem exists, at least I have a certain sender to whom replying crashes my v4.80 Pmail. IErenderer updated. Any solution to this? Thanks in advance. Peter


Hi, Still today it seems this problem exists, at least I have a certain sender to whom replying crashes my v4.80 Pmail. IErenderer updated. Any solution to this? Thanks in advance. Peter

What makes you think it's the same issue? Crashing on reply can have lots of reasons, so without taking a look into such a particular message nobody can answer this question, and without knowing about the details we can't provide a solution (or at least work-around), so either you provide the details for what exactly in detail causes the issue or provide a sample message for us to figure it out (the latter being preferred, of course): If possible forward it as an attachment (this is important) to <beta-reports [at] pmail.gen.nz> using a subject referring to this thread here so we can tell it apart from all the spam getting sent to this account.


What you can also try for replies, though, is to highlight the whole message (Ctlr + A) and reply to the selection, this may help as well as wrapping or reformatting (if it's the same issue), but we would still be interested in the exact reason for crashing.


TIA


What makes you think it&#039;s the same issue? Crashing on reply can have lots of reasons, so without taking a look into such a particular message nobody can answer this question, and without knowing about the details we can&#039;t provide a solution (or at least work-around), so either you provide the details for what exactly in detail causes the issue or provide a sample message for us to figure it out (the latter being preferred, of course): If possible forward it _as an attachment_ (this is important) to &lt;beta-reports [at] pmail.gen.nz&gt; using a subject referring to this thread here so we can tell it apart from all the spam getting sent to this account. What you can also try for replies, though, is to highlight the whole message (Ctlr + A) and reply to the selection, this may help as well as wrapping or reformatting (if it&#039;s the same issue), but we would still be interested in the exact reason for crashing. TIA
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C
edited Apr 28 '23 at 11:14 pm

If possible forward it as an attachment


Got your test message and it doesn't crash on my machine and doesn't even show anything unusual when running in a debugger even with long lines enabled. Try updating to the current beta v4.81 from https://www.pmail.com/v48x.htm.


[quote=&quot;pid:55391, uid:2133&quot;]If possible forward it as an attachment[/quote] Got your test message and it doesn&#039;t crash on my machine and doesn&#039;t even show anything unusual when running in a debugger even with long lines enabled. Try updating to the current beta v4.81 from https://www.pmail.com/v48x.htm.
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C
edited Apr 29 '23 at 6:39 pm

I am happy to test it on 4.80 and 4.81 if that could provide any enlightenment.


I am happy to test it on 4.80 and 4.81 if that could provide any enlightenment.

I am happy to test it on 4.80 and 4.81 if that could provide any enlightenment.

Brian confirmed that the issue has been fixed by David Harris in the public beta v4.81 while it still occurs in v4.80.


[quote=&quot;pid:55393, uid:28772&quot;]I am happy to test it on 4.80 and 4.81 if that could provide any enlightenment.[/quote] Brian confirmed that the issue has been fixed by David Harris in the public beta [v4.81](https://www.pmail.com/v48x.htm) while it still occurs in v4.80.
			Michael
--
IERenderer's Homepage
PGP Key ID (RSA 2048): 0xC45D831B
S/MIME Fingerprint: 94C6B471 0C623088 A5B27701 742B8666 3B7E657C

More specifically, it only occurs on 4.80 when you select "None" as the reply option for "Special handling for long lines from the original message". There is no crash when you select "Reformat" or "Wrap".


More specifically, it only occurs on 4.80 when you select &quot;None&quot; as the reply option for &quot;Special handling for long lines from the original message&quot;. There is no crash when you select &quot;Reformat&quot; or &quot;Wrap&quot;.

Hi,


Thanks much for all your replies, it's clear now.


Best,
Peter


Hi, Thanks much for all your replies, it&#039;s clear now. Best, Peter
live preview
enter atleast 10 characters
WARNING: You mentioned %MENTIONS%, but they cannot see this message and will not be notified
Saving...
Saved
With selected deselect posts show selected posts
All posts under this topic will be deleted ?
Cancel
Delete
Pending draft ... Click to resume editing
Discard draft