home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.unix.questions:9265 comp.unix.sysv386:12223 comp.unix.programmer:3839 comp.unix.wizards:3274
- Newsgroups: comp.unix.questions,comp.unix.sysv386,comp.unix.programmer,comp.unix.wizards
- Path: sparky!uunet!cs.utexas.edu!csc.ti.com!tilde.csc.ti.com!m2.dseg.ti.com!ernest!alan
- From: alan@ernest.dseg.ti.com (Alan Edmonds)
- Subject: Re: How to tell which process(es) have a given file/inode open under SVr4?
- Organization: Texas Instruments, Inc. - Plano, Tx
- Date: Wed, 22 Jul 1992 00:38:22 GMT
- Message-ID: <BrrLrz.MxI@ernest.dseg.ti.com>
- Keywords: File Process Inode Open SVR4
- References: <1992Jul21.163221.1417@mjbtn.jobsoft.com>
- Lines: 73
-
- In article <1992Jul21.163221.1417@mjbtn.jobsoft.com> root@mjbtn.jobsoft.com (Mark J. Bailey [ADMIN]) writes:
- >Hello,
- >
- >Sorry for the multi-group crosspost. I was not sure where to look.
-
- You got close enough.
-
- [some stuff deleted]
- >OK history aside, what is now happening is that the initial parameters
- >on the ports in question are 9600 (and so on), ie, not what you would
- >expect, and after our open call and ioctl calls, NOTHING BLOODY CHANGES
- >at all, our program output gets post processed (because for some reason
- >OPOST [along with everything else] that we clear [which worked on the
- >ports never having had terminal service] in not getting cleared), and
- >things are a mess.
-
- I'm not sure I understand your problem. But here goes anyway.
-
- This is normal behavior. The ports will revert to what the kernal
- wants them to be (300 baud, etc), when the LAST open terminates.
- SO, you can open the port, dink with it, but when you close it BAM,
- back to 300 baud. The "trick" is to keep the port open with
- something like "sleep 999999 </dev/ttynn &". This will keep the port
- open so when your setup terminates, the changes will keep. You can
- test this by trying to change the port with stty and see if the changes
- take. Then do the same changes after running the sleep command above.
-
-
- >I REMOVED ports services for the ports in question and no change. We
- >contacted the software vendor for the other application that is also on
- >this system, and to their knowledge, there is nothing in their package
- >that would go out and open these ports AND KEEP THEM OPEN. It is this
- >concept of something, somewhere, that is opening these ports and
- >keeping them open (or somehow locked) (and thus preventing our ioctl's
- >from having any effect) that we are now focusing on. This is also
- >supported by the fact that initial port parms should have a baud rate
- >of 300. On the ORIGINAL ports we used, as I said, they were initially
- >300 baud, went to 9600 after our inits, then at close, immediately went
- >back to 300 baud.
-
- Stupid question, but I have to ask; What about getty? ttymon?
-
- >This has turned into a longwinded hurl and I apologize! :-) Anyway,
- >I would think that if a port had *NO* port service DEFINED (not just
- >DISABLED), and no other process had it opened, it should be 300 baud.
- >So, something has to have it and we are not sure how to tell whom. Under
- >SVr3 (SCO), the pstat command aided in locating open inodes and their
- >opening processes. But (and here is the real reason for my post), how
- >do you do it under SVr4??????? :-) And, maybe it is not what I am
- >leaning towards at all and if you have another thought (anything! :-)),
- >we would be most appreciative to you! If SVr4 was in house and used
- >everyday here, no biggie, I am sure, but this is one of those "HHhhmmm....
- >how did I manage to get in this #^%#&^% mess?" situtations and we are
- >stumped.
- >
- >Thank you for your time and endurance!
- >
- >Mark
- >
- >--
- >Mark J. Bailey, N4XHX _______/====X11====\_______
- >USMAIL: 511 Memorial Blvd., Murfreesboro, TN 37129 | JobSoft |
- >VOICE: +1 615 893 0098 | Design & Development Co.|
- >UUCP: ...!uunet!mjbtn!mjb, ...!raider!mjbtn!mjb | Murfreesboro, TN USA |
- >DOMAIN: mjb@mjbtn.JOBSOFT.COM CIS: 76314,160 ---------------------------
- ><KA9Q-UNIX-USERS Mailing List-Subscribe: ka9q-unix-requests@mjbtn.jobsoft.com>
-
-
- --
- Alan Edmonds Texas Instruments, Inc.
- I don't speak for TI; TI doesn't speak for me M/S 8513
- Work phone: (214)575-6427 6620 Chase Oaks Blvd.
- Email: alan@ernest.dseg.ti.com Plano, Texas 75023
-