rickc@telly.on.ca (Rick Copley) (01/22/91)
Can anyone out there tell me how to tell if a given process is actually doing any thing (ie accumulating cpu time) from within a C program ?? On this NCR TOWER 32/800 running AT&T s5r3 (hacked by NCR, 2.00.02 is the NCR version number), I have found a file called /usr/include/sys/proc.h that seems to have all the stuff I need to "monitor" a process, but I can't figure out what "function" to call to actually use it. Reason I am looking is we have a problem with users NOT login off, and causing swappage for the other users. They are the kind that login, type for 20 seconds and leave the terminal logged in for hours and hours. I have taken a C program from UnixWorld Mag that looks at the users /dev/tty file to see when the last time they TYPED was. This is no go for when a user runs a very lengthy report that does no screen I/O. The C program (called autologout.c) murders EVERYONE who does not access his/her tty port for more than a certain amount of time. I realize I could muck around with the ps -t command and capture the cpu time in a file and compare it to a few minutes ago and stuff like that, but it seems like there should be a REALLY easy way to check if a process is "running" and accumulating cpu time. Help, anyone ??? -- rickc@telly.on.can "You can twist perception, but reality wont budge" - RUSH uunet!telly.on.ca!rickc (Rick Copley)
pfalstad@phoenix.Princeton.EDU (Paul Falstad) (01/22/91)
rickc@telly.on.ca (Rick Copley) wrote: >Can anyone out there tell me how to tell if a given process is actually >doing any thing (ie accumulating cpu time) from within a C program ?? > >I realize I could muck around with the ps -t command and capture the cpu time >in a file and compare it to a few minutes ago and stuff like that, but it seems >like there should be a REALLY easy way to check if a process is "running" and >accumulating cpu time. Yes, there should be, but there isn't. You need to read /vmunix's name list, open /dev/kmem, and traverse the kernal process table. It's not pretty at all. Reading the output of some variant of ps is probably the easiest solution in this case. -- Paul Falstad, pfalstad@phoenix.princeton.edu PLink:HYPNOS GEnie:P.FALSTAD In the heat of composition I find that I have inadvertently allowed myself to assume the form of a large centipede. I am accordingly dictating the rest to my secretary.
mike (Michael Stefanik) (01/23/91)
In article <279B5779.58AF@telly.on.ca> telly.on.ca!rickc (Rick Copley) writes: >Can anyone out there tell me how to tell if a given process is actually >doing any thing (ie accumulating cpu time) from within a C program ?? >[...] >I realize I could muck around with the ps -t command and capture the cpu time >in a file and compare it to a few minutes ago and stuff like that, but it seems >like there should be a REALLY easy way to check if a process is "running" and >accumulating cpu time. Unfortunately, this is one of the significant failings of UNIX ... it doesn't provide any kernel function that reads the namelist on behalf of a process. It isn't that reading the namelist is so tough either ... the problem is that every flavor of UNIX loves to give it a different symbol name. :-( IMHO, there should be a function called procstat() that returns a pointer to an array of 'proc' structs. Of course, it'll never happen. -- Michael Stefanik, Systems Engineer (JOAT), Briareus Corporation UUCP: ...!uunet!bria!mike -- technoignorami (tek'no-ig'no-ram`i) a group of individuals that are constantly found to be saying things like "Well, it works on my DOS machine ..."
emv@ox.com (Ed Vielmetti) (01/23/91)
In article <279B5779.58AF@telly.on.ca> rickc@telly.on.ca (Rick Copley) writes:
Can anyone out there tell me how to tell if a given process is actually
doing any thing (ie accumulating cpu time) from within a C program ??
pmon does this. it runs at least on decs, look on gatekeeper. dunno
about system v.
--Ed
shore@mtxinu.COM (Melinda Shore) (01/24/91)
In article <382@bria> bria!mike (Michael Stefanik) writes: >IMHO, there should be a function called procstat() that returns a pointer to >an array of 'proc' structs. Of course, it'll never happen. That's not very general. There are many kernel data structures that are interesting/useful/whatever, and you shouldn't need a different system call for each one. Instead, you can use something like the table system call (which Mach HAS) or the tabinfo/tabread system calls (which Unicos HAS). Also, from within the kernel you don't have to read the namelist, because, of course, addresses are known at link time. And names of kernel data structures don't vary that much within major Unix strains (or even among them - is there a Unix out there where the name of the proc table isn't "proc"?). You'll find sysinfo in just about any sysV variant there is, for example. [BTW, I would have sent you email about this but your "Reply-to:" line totally loses and you have no From: line in your header.] -- Software longa, hardware brevis Melinda Shore shore@mtxinu.com mt Xinu ..!uunet!mtxinu.com!shore