mike@bria.AIX (Mike Stefanik/78125) (12/28/90)
I having a bit of a problem with an RS/6000 and a modem on the
S2 serial port. I'm running 3.1 (3002 updates), and have a
Multitech 224E sitting on the S2 port (and, of course, using a
standard IBM modem cable for said port). The port is the standard
2400 bps, 81N, and *enabled* for login (not shared; that doesn't seem
to work *at all*, even with 3002 applied).
The problem is that when I try to dial up the system, I get a
connect, but no login. Now, what is interesting is that *infrequently*
when two "pdisable" and two "penable" commands are issued, such as:
# pdisable tty1; pdisable tty1; penable tty1; penable tty1
I have gotten a login, and from that point on was able to login
consistently. However, once the machine is rebooted, it goes on the
fritz again.
BTW: There is a getty on the port at the time ...
Anyone out there have any ideas?
-----------------------------------------------------------------------------
Michael Stefanik, Systems Engineer (JOAT), Briareus Corporation
UUCP: ...!uunet!bria!mike
"If it was hard to code, it should be harder to use!"ghe@nucthy.physics.orst.edu (Guangliang He) (12/31/90)
In article <281@bria.AIX> mike@bria.AIX (Mike Stefanik/78125) writes: > I having a bit of a problem with an RS/6000 and a modem on the > S2 serial port. I'm running 3.1 (3002 updates), and have a > Multitech 224E sitting on the S2 port (and, of course, using a > standard IBM modem cable for said port). The port is the standard > 2400 bps, 81N, and *enabled* for login (not shared; that doesn't seem > to work *at all*, even with 3002 applied). > > The problem is that when I try to dial up the system, I get a > connect, but no login. I have a similar situation here. I have a Multitech modem connect to one of the 8 port adapter (not S2). Login is enabled. I can get connect 2400 but no login prompt. My solution is hit Ctcl-C or Ctcl-\ after I got the connect. Works for me everytime. Guangliang He ghe@PHYSICS.ORST.EDU hegl@ORSTVM.BITNET
mike@bria.AIX (Mike Stefanik/78125) (01/01/91)
In article <22165@orstcs.CS.ORST.EDU>, ghe@nucthy.physics.orst.edu (Guangliang He) writes: > In article <281@bria.AIX> mike@bria.AIX (Mike Stefanik/78125) writes: > > I having a bit of a problem with an RS/6000 and a modem on the > > S2 serial port. I'm running 3.1 (3002 updates), and have a > > Multitech 224E sitting on the S2 port (and, of course, using a > > standard IBM modem cable for said port). The port is the standard > > 2400 bps, 81N, and *enabled* for login (not shared; that doesn't seem > > to work *at all*, even with 3002 applied). > > > > The problem is that when I try to dial up the system, I get a > > connect, but no login. > > I have a similar situation here. I have a Multitech modem connect to > one of the 8 port adapter (not S2). Login is enabled. I can get connect > 2400 but no login prompt. My solution is hit Ctcl-C or Ctcl-\ after I > got the connect. Works for me everytime. When I called AIX support, that was one of the things that they suggested that I do; I've tried CTRL-C, DEL, and BREAK and nothing seems to get that modem's attention. Unfortunately, there were two APARs that they dug up related to this, and as I gather, they both were included in the 3002 update. Now, it looks as though they've dropped the ball on me, and I have a customer who I can't support. ----------------------------------------------------------------------------- Michael Stefanik, Systems Engineer (JOAT), Briareus Corporation UUCP: ...!uunet!bria!mike "If it was hard to code, it should be harder to use!"
robin@batcomp.austin.ibm.com (Robin D. Wilson) (01/03/91)
In article <300@bria.AIX> mike@bria.AIX (Mike Stefanik/78125) writes: >In article <22165@orstcs.CS.ORST.EDU>, ghe@nucthy.physics.orst.edu (Guangliang He) writes: >> In article <281@bria.AIX> mike@bria.AIX (Mike Stefanik/78125) writes: >> > I having a bit of a problem with an RS/6000 and a modem on the (stuff deleted) >> > The problem is that when I try to dial up the system, I get a >> > connect, but no login. > >When I called AIX support, that was one of the things that they suggested >that I do; I've tried CTRL-C, DEL, and BREAK and nothing seems to get >that modem's attention. There should be no APARs that are related to your problem. (Assuming I know what the problem really is.) There are several APARs relating to serial ports and SHARE/DELAY not working. The reason "^C" or "BREAK" might work is because they will cycle the getty on the port and cause it to respawn. ("^C" will kill the getty, BREAK/DEL will cause a framing error, forcing getty to respawn.) When it respawns, it opens the port cleanly, and you can login.... This doesn't work probably because your port has been listed as defective by the OS since it was attempting to login over and over again until the system figured that it was defective and disabled it. The reason this happens is 2 fold: 1) the modem is setup wrong; that is... it is setup to respond with "RING" everytime someone calls in, and carrier is likely strapped high on the modem switch settings (either software or hardware switches). When you start a getty (PENABLE) on the port, it locks the port andthen sits there waiting for carrier. When carrier is detected, getty calls the "logger" (normally/default == /bin/login) which issues the login herald. When the modem has carrier strapped high, getty instantly goes to the logger, which sends the herald. The modem assumes that the herald is an attempt to issue a command, but it doesn't recognize the command, so it says "ERROR". The logger thinks "USER ERROR IS ATTEMPTING TO LOGIN" and sends "password:" to which the modem responds "ERROR", to which getty restarts, and everything starts over again. After several hundred iterations, the port is marked as defective, and getty is disabled. 2) The port is set to "ENABLE". I realize that at this point, SHARE and DELAY are broken, so proper tty setup cannot be realized; but for future referance (and 3003 update levels): the port will assume that a device (aka. terminal) is directly attached, and will not really handle all of the modem idiosyncraies very well. If you still have a problem with this, give me a call at the number below.... Please don't expect a great deal of "how-to" assistance, but I can give you the basic run-down on the tty problems. -- +-----------------------------------------------------------------------------+ |The views expressed herein, are the sole responsibility of the typist at hand| +-----------------------------------------------------------------------------+ |UUCP: robin%aixserv@uunet.uu.net | |USNail: 701 Canyon Bend Dr. | | Pflugerville, TX 78660 | | Home: (512)251-6889 Work: (512)823-3015 | +-----------------------------------------------------------------------------+
jfh@greenber.austin.ibm.com (John F. Haugh II) (01/08/91)
In article <281@bria.AIX> mike@bria.AIX (Mike Stefanik/78125) writes: >The problem is that when I try to dial up the system, I get a >connect, but no login. Now, what is interesting is that *infrequently* >when two "pdisable" and two "penable" commands are issued, such as: > > # pdisable tty1; pdisable tty1; penable tty1; penable tty1 There have been some problems related to pdisable/penable versus using SMIT. I'd suggest trying the SMIT command to enable a TTY for login. You may want to check the contents of your /etc/inittab file to see if the penable command is putting the correct line in that file. Also, you need to get PTF 3003 as it includes fixes to other commands as well. -- John F. Haugh II | This space intentionally | MaBellNet: (512) 838-4330 SneakerNet: 809/1C079 | left blank ... | VNET: LCCB386 at AUSVMQ BangNet: ...!cs.utexas.edu!ibmchs!auschs!snowball.austin.ibm.com!jfh (e-i-e-i-o)
mike@bria.UUCP (Michael Stefanik) (01/09/91)
In article <4687@awdprime.UUCP> John F. Haugh II writes: >There have been some problems related to pdisable/penable versus >using SMIT. I'd suggest trying the SMIT command to enable a TTY >for login. As I understand it, SMIT is just an interface for the regular commands such as penable, pdisable, and chdev. Would there really be a functional difference if I enter "penable" or SMIT enters "penable"? >You may want to check the contents of your /etc/inittab file to >see if the penable command is putting the correct line in that >file. Also, you need to get PTF 3003 as it includes fixes to >other commands as well. Yup, inittab looked wonderful, and ps does show a getty on the port. From the looks of things, I'll have to snag the 3003 update and see if that fixes the problem ... :-( -- 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 ..."
jfh@greenber.austin.ibm.com (John F Haugh II) (01/10/91)
In article <334@bria> mike@bria.UUCP (Michael Stefanik) writes: >As I understand it, SMIT is just an interface for the regular commands >such as penable, pdisable, and chdev. Would there really be a functional >difference if I enter "penable" or SMIT enters "penable"? Generally speaking, yes, SMIT is just an interface to the usual collection of commands used to alter a device. However, in this specific instance SMIT does not use "penable" or "pdisable" to do its work, instead it calls "chdev -l tty0 -a login=<>" or something to that effect. If you are really interested, call up SMIT and make the changes you desire, then press the <F6> key instead of <RETURN>. That will tell you exactly what commands SMIT is going to execute. >Yup, inittab looked wonderful, and ps does show a getty on the port. >From the looks of things, I'll have to snag the 3003 update and >see if that fixes the problem ... :-( There are some fixes in 3003 that will fix problems with shared ports (-u and -r and -??? flags). Your problem sounds like something completely different that is also fixed in 3003 that related to p{delay,share,enable,disable}. -- John F. Haugh II | This space intentionally | MaBellNet: (512) 838-4330 SneakerNet: 809/1C079 | left blank ... | VNET: LCCB386 at AUSVMQ BangNet: ...!cs.utexas.edu!ibmchs!auschs!snowball.austin.ibm.com!jfh (e-i-e-i-o)