XYplorer Beta Club

Hi
I am getting at least 15 crashes a day, something that never ever happened with previous version(single pane). The crash is not like a crash with a message, it literally disappears from my desktop. I do not want to be negative about the development but dual pane period brought quite a bit of frustration in the stability area. In last 10 min Xyplorer just disappeared 3 times. I am not even using any right click or anything. Sometimes I change to another window and boom, Xyplorer is not there anymore. Since there is not any debug mode I cannot provide you any more information beside telling my experience.

I have also started the massive startup delay issue under another topic. That was resolved before but now even with the suggested tweaks the startup is very very slow some 20k milliseconds.

As a way to narrow things down, backup all your settings files and then completely reset XY to factory defaults and see if you get the same issues - this way we can narrow it down to being a direct program problem, or some combination of customised settings.

i'm getting like that too...
random crash, even not doing anything (strange)
i get about 6 crashes in the last 2 days

i cant remember it, i think the last 2 or so release

anyway, i'm going back to work it out (its going to be awhile)

BTW i'm on x86

The topic is "a lot of crashes undr xp64". This appears to imply that the crashes are somehow connected to xp64 resp. x64 OS in general. Please note that this is not possible*, and therefore please don't put such implications into the topic lines because it might lead the casual forum visitor to think that XY somehow does not work under x64 systems! Thanks.

* It's not possible because 32-bit apps do of course work perfectly on 64-bit systems (otherwise MS would have no chance to sell 64 OSs). And if they would be any problem it would be the fault of the OS, not of the app.

I changed the topic line to "a lot of crashes undr xp". ("under XP" again is an implication but at least one that might be valid)

Conc. the crashes. Show me a way to reproduce one.

admin wrote:Conc. the crashes. Show me a way to reproduce one.

I think that there is something fishy about Vpn folder tabs+Xyplorer. I got rid of all my Vpn tabs for tests and crashes stopped (so far). I do not know if this would help you but that is the best I could come up with for now.

Since the Jul 8th beta release I am getting unexplained crashes as well. I cannot reproduce it - but for some reason 3 or 4 times a day with XYPlorer in the background I XYplorer crashes and has to exit. Most often this happens with XYPlorer in the background and I am browsing the web or writing a Word doc or something.

I know it is not a helpful description but it is all I have - this product used to NEVER crash for me and now it is. I will do all I can to investigate to see if there is some conflict on my end but I wanted to post this because I see other user's having unexplained crashes. I am using WinXP PRO Sp2 32 bit with loads of ram and processor speed.

Ok, we have 3 users reporting random crashes since about 1 month ago, and the crashes are happening when XY is in the background. OS is XP.

I can conclude at least the following:
1) crashes depend on certain user settings AND/OR certain OS settings -- else more users would report such problems
2) crashes are related to auto-refresh, otherwise XY is totally inactive when in the background and would never crash

Not much, and not enough. Please try to narrow it down somehow.

PS: I did some voodoo anyway so tell me what's happening to those crashes in the next beta...

I know this is probably a unique situation, but it might give another pointer to something happening with Auto-Refresh.

Files are being sent to me. A header is sent first which contains the file name and size. This creates the file on disk and allocates the required disk space. An acknowledge is then sent to the sender and the file transfer starts. This is code I wrote myself, but is fairly standard protocol. In XY sometimes the file name is displayed but with zero size, which then changes to the correct size when the first block of the file is written to disk. Because all the headers are sent together, before the file transfer starts, several files can sometimes have zero space displayed in XY.

It doesn't happen in x² or T.C watching the same directory simultaneously in either x² or T.C and XY.

Gandolf wrote:I know this is probably a unique situation, but it might give another pointer to something happening with Auto-Refresh.

Files are being sent to me. A header is sent first which contains the file name and size. This creates the file on disk and allocates the required disk space. An acknowledge is then sent to the sender and the file transfer starts. This is code I wrote myself, but is fairly standard protocol. In XY sometimes the file name is displayed but with zero size, which then changes to the correct size when the first block of the file is written to disk. Because all the headers are sent together, before the file transfer starts, several files can sometimes have zero space displayed in XY.

It doesn't happen in x² or T.C watching the same directory simultaneously in either x² or T.C and XY.

XY just reports what it is told by Windows. Maybe XY has a faster refresh rate than the other managers so they don't see the zero size state?

Possible. I just checked and XY's list updates before x² does but it's not easy to see if there is any difference between XY and T.C. - XY might be microseconds faster.

So, is XY seeing the file creation from Windows before the space is allocated, and then not flagging the change in disk allocation as a change? Hence, the error might occur on larger files and if the allocated space is fragmented. I'll keep a check out of curiosity.

But, also, under certain conditions, XY might be getting incorrect information about file system changes, which is causing these crashes. I haven't been aware of any unexplained crashes, I can usually tie them down to something causing them.

I can change the refresh rate in x² because it's length has been discussed on the forum. I don't know about T.C. but I will look.

Gandolf wrote:Possible. I just checked and XY's list updates before x² does but it's not easy to see if there is any difference between XY and T.C. - XY might be microseconds faster.

So, is XY seeing the file creation from Windows before the space is allocated, and then not flagging the change in disk allocation as a change? Hence, the error might occur on larger files and if the allocated space is fragmented. I'll keep a check out of curiosity.

But, also, under certain conditions, XY might be getting incorrect information about file system changes, which is causing these crashes. I haven't been aware of any unexplained crashes, I can usually tie them down to something causing them.

I can change the refresh rate in x² because it's length has been discussed on the forum. I don't know about T.C. but I will look.

mikes16 has confirmed in the meantime that the unexplained background crashes have been fixed in v8.10.0020. They were existing for exactly one month in the BETAs after 8.10.0000.

Gandolf, I did not read your post as a bug report. To me it looks like XY behaves OK. Where do you see a bug here?

As i notice a lot of hardly reproducible crashes under xp, too: Would it help to supply dmp files of the crashes? Sysinternal's ProcDump can produce them.

ogroeger wrote:As i notice a lot of hardly reproducible crashes under xp, too: Would it help to supply dmp files of the crashes? Sysinternal's ProcDump can produce them.

Tried v8.10.0020?

Tried v8.10.0020?

Not yet, but i'll download it now.