| TIP: Click on subject to list as thread! | ANSI |
| echo: | |
|---|---|
| to: | |
| from: | |
| date: | |
| subject: | Monitor DCD status |
Hello Craig! 24 Aug 94 20:35, Craig Morrison wrote to Marcello Desantis: CM> I can't think of any reason why you would need to have a CM> watchdog running with Bink... I wrote that Binkley prevents me from opening the com port, but the one I need to monitor is Maximus and not Binkley Term. We are experiencing frequent hangs when Maximus spawns external programs. Most of the times the problem regards MAXPIPE running zip (InfoZip) to compress QWK packets, but even os2you, under some circumstances, doesn't detect carries losses and it doesn't return control to Maximus. So I had the simple idea to monitor, from another process, both the binkley/maximus log pipe and the DCD status. If the carrier goes down while an external program is running I wait, say, 30 seconds for Maximus to regain control and, if it doesn't, I first try to kill the offending process and then, if I can't, reboot. CM> Accept the port handle as a parameter on your CM> command line, start a thread to watch the DCD state and then CM> spawn the next executable in your main thread passing the port CM> handle along.. In this case you don't have to worry about opening CM> and closing the port, just convert it from ASCII to an HFILE and CM> call DosDevIOCtl with it. Yes. I may get Binkley spawning my process and the latter spawning Maximus. But I'd prefer to have the job done from an indipendent process to avoid delaying Maximus startup. Eventually, I'd rather let my process spawn Binkley Term itself. Anyway, if I have no alternatives, I'm gonna implement your method. The BBS is unattended and 40 kilometers away from my house and I prefer to have our users waiting 2 seconds more than driving all across the town to restart the system. MD>> b) Reboot the system if some unrecoverable error occur. I do CM> This one I can't help you with.. And I'm sure you are going to CM> be asked by a lot of folks why you think this is really CM> necessary.. I agree that rebooting SHOULD never be necessary, but in real life the only system I've seen that has not been requering a reboot for over a year was a Stratus fault-tolerant machine that I've been working on several years ago (it was the authorization machine for an italian CC issuer). To come back to my scenario, I've seen cases where DosKillProcess doesn't work and I'd like to have rebooting acting as a "last chance" action. Marcello --- GoldED 2.42.G0214+* Origin: Hold the line (2:335/350) SEEN-BY: 12/2442 54/54 620/243 624/50 632/348 640/820 690/660 711/409 410 413 SEEN-BY: 711/430 807 808 809 934 712/353 623 713/888 800/1 @PATH: 335/350 336 331/334 301/1 24/24 396/1 3615/50 229/2 @PATH: 12/2442 711/409 54/54 711/808 809 934 |
|
| SOURCE: echomail via fidonet.ozzmosis.com | |
Email questions or comments to sysop@ipingthereforeiam.com
All parts of this website painstakingly hand-crafted in the U.S.A.!
IPTIA BBS/MUD/Terminal/Game Server List, © 2025 IPTIA Consulting™.