Re: [WebDNA] sem_open solved
This WebDNA talk-list message is from 2012
It keeps the original formatting.
numero = 108302
interpreted = N
texte = On Jan 26, 2012, at 2:58 PM, Govinda wrote: > >> I'd like to personally thank Chris and Josh for all their help on this one, and Thierry for his part in getting it going and splitting the costs of the executable. >> >> Just for anyones info, it was a 2 (maybe 3) part problem. >> >> 1) The executable itself >> 2) The fact that you have to restart a server after 210 days(who knew... it's a server...sheeesh... lol ) >> 3) Maybe bad ram contributing to it, but thats a wait and see moment. >> >> Prior servers to this have run for over a year without restart, but I understand the reasoning for this, as everything related to semaphores was all heaped up. Sometimes you get so wrapped up in this crap, you don't think of things like that, especially when everything else seems to run fine. >> >> Anyway, its fixed, and thank you to everyone for their input and help. > > Rob > > Sorry if I am being foggy.. but I did not understand very well.. and am curious. So say again, what exactly was causing your crashes? And you solved it? How exactly? ...primarily by paying WSC for a new version of 6.x , minus the semaphore bug? > > -Govinda It was a combination of the bug and heaping in the system and the semaphores that was crashing webcatalog. I'm not exactly sure, but buying the executable probably stopped the heaping, I would assume, because it was continually calling for them. The problem was aggravated by me not restarting it for long periods of time, so it just kept heaping up, I would imagine. Restarting it probably would have returned everything back to normal for a specific period of time until it would heap up again. I'm assuming here, but that seems where it was going, and it would cause the webcatalog process to run up to max until the process was restarted, which would last for a period, depending on the hits on the server. Thierry was having sem_open issues as well, probably not as severe cause he probably restarted every few months or so, so him and I split the cost of paying for the development of an executable to correct the problem to WSC. Not the whole program, but just the WebCatalog executable inside. This was on the Mac Os x version of 6.2. Rob
Associated Messages, from the most recent to the oldest:
On Jan 26, 2012, at 2:58 PM, Govinda wrote: > >> I'd like to personally thank Chris and Josh for all their help on this one, and Thierry for his part in getting it going and splitting the costs of the executable. >> >> Just for anyones info, it was a 2 (maybe 3) part problem. >> >> 1) The executable itself >> 2) The fact that you have to restart a server after 210 days(who knew... it's a server...sheeesh... lol ) >> 3) Maybe bad ram contributing to it, but thats a wait and see moment. >> >> Prior servers to this have run for over a year without restart, but I understand the reasoning for this, as everything related to semaphores was all heaped up. Sometimes you get so wrapped up in this crap, you don't think of things like that, especially when everything else seems to run fine. >> >> Anyway, its fixed, and thank you to everyone for their input and help. > > Rob > > Sorry if I am being foggy.. but I did not understand very well.. and am curious. So say again, what exactly was causing your crashes? And you solved it? How exactly? ...primarily by paying WSC for a new version of 6.x , minus the semaphore bug? > > -Govinda It was a combination of the bug and heaping in the system and the semaphores that was crashing webcatalog. I'm not exactly sure, but buying the executable probably stopped the heaping, I would assume, because it was continually calling for them. The problem was aggravated by me not restarting it for long periods of time, so it just kept heaping up, I would imagine. Restarting it probably would have returned everything back to normal for a specific period of time until it would heap up again. I'm assuming here, but that seems where it was going, and it would cause the webcatalog process to run up to max until the process was restarted, which would last for a period, depending on the hits on the server. Thierry was having sem_open issues as well, probably not as severe cause he probably restarted every few months or so, so him and I split the cost of paying for the development of an executable to correct the problem to WSC. Not the whole program, but just the WebCatalog executable inside. This was on the Mac Os x version of 6.2. Rob
Rob
DOWNLOAD WEBDNA NOW!
Top Articles:
Talk List
The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...
Related Readings:
I get a lot of spam (2002)
[WebDNA] To be or not to be friendly (URLs) (2011)
OSX Server & WebStar5 Software 4Sale (2003)
[Announce] Newest Commerce Site based on WebCatalog (1997)
Cancel Subscription (1996)
WebCatalog/Mac 2.1b2 New Features (1997)
User/pass with tcpconnect (2000)
[WebDNA] Cannot get WebDNA and MySQL working. (2015)
[Middle] Probs with "&" (2003)
[append] vs. [appendfile] delta + question? (1997)
OS Limitations (1998)
PCS Frames (1997)
Bug or syntax error on my part? (1997)
Web Site (2000)
Modifying Carts (1999)
Banners and sort of random display (1997)
Images (2000)
Dumb Question about Docs (1997)
What about this [delete] situation? (1998)
WebDNA-Talk Digest mode broken (1997)