|
View previous topic :: View next topic
|
| Author |
Message |
vasanthz
Global Moderator
Joined: 28 Aug 2007
Posts: 1744
Location: Tirupur, India
|
|
|
|
Hi,
I am trying to configure SSHD on Mainframe using dovetail.com/docs/pt-quick-inst-12/pt-quick-inst-12-doc.pdf
I started the SSHD daemon using /S SSHD
It started many processes like SSHD1, 2, 3, 4 etc.. Currenlty I have only one process active called SSHD5 on the SDSF "PS" screen.
| Code: |
SDSF PROCESS DISPLAY SYSTEMA ALL INVALID COMMAND
COMMAND INPUT ===> SCROLL ===> CSR
PREFIX=SSH* DEST=(ALL) OWNER=* SYSNAME=
NP JOBNAME JobID Status Owner State CPU-Time
SSHD5 STC00100 SWAPPED,FILE SYS KERNEL WAIT USERID 1FI 0.03
|
When I try to connect to port 22 using Putty "putty userid@host" it says "Network Error: Software caused connection abort"
Could you please let me know where we can find the error messages pertaining to this connection attempt? and how to proceed with debugging
Below is the SSHD proc
| Code: |
//SSHD PROC
// EXEC PGM=BPXBATCH,REGION=0M,TIME=NOLIMIT,
// PARM='PGM /bin/sh -c /etc/ssh/sshd.sh'
//STDERR DD SYSOUT=* |
Regards,
Vasanth.S |
|
| Back to top |
|
 |
Robert Sample
Global Moderator
Joined: 06 Jun 2008
Posts: 8700
Location: Dubuque, Iowa, USA
|
|
|
|
| Did you look in /tmp/syslogd.log (or whatever you called syslogd)? Section 1.13 of the manual you linked in your post talks about setting up syslogd for SSHD. It also explicitly states that SSHD does not write to the console if syslogd is not available. |
|
| Back to top |
|
|
vasanthz
Global Moderator
Joined: 28 Aug 2007
Posts: 1744
Location: Tirupur, India
|
|
|
|
Thanks Robert, Even on the slimmed down installation I cut corners by not paying attention to SYSLOGD part. This is sort of a proof of concept install and not the actual production install.
I don't know what SYSLOGD is, I'll look it up and see if we have it setup. |
|
| Back to top |
|
|
Robert Sample
Global Moderator
Joined: 06 Jun 2008
Posts: 8700
Location: Dubuque, Iowa, USA
|
|
|
|
| syslogd is the syslog daemon for Unix System Services. See the IP Configuration Guide manual in the Communications Server bookshelf for details on how to configure it. For a quick check, look at /etc/syslog.conf to see if this file exists; if not, then syslogd is either not running or running entirely with defaults (which is rarely good). If it is not there, copy over /usr/lpp/tcpip/samples/syslog.conf and customize it to your site. |
|
| Back to top |
|
|
vasanthz
Global Moderator
Joined: 28 Aug 2007
Posts: 1744
Location: Tirupur, India
|
|
|
|
Thanks for the help again Robert, the suggestion about SYSLOGD helped resolving the issue.
SYSLOGD was not running. I changed the SYSLOGD proc and started it.
It throws in a lot of messages on the SYSLOG, but it appears to work for now. Need to find a way to limit the number of messages it writes.
There was an error whenever I tried to connect using Putty on the syslog,
| Code: |
ICH408I USER(SSHXX ) GROUP(XXXX ) NAME(SSHD PRIVILEGE SEPAR) 453
LOGON/JOB INITIATION - REVOKED USER ACCESS ATTEMPT |
Resumed the ID and now we are good to go. SSH works! :-)
I feel like I have awoken Frankenstein :-)
Regards,
Vasanth.S |
|
| Back to top |
|
|
Robert Sample
Global Moderator
Joined: 06 Jun 2008
Posts: 8700
Location: Dubuque, Iowa, USA
|
|
|
|
I'm glad you have it working now!
It is pretty common to have syslogd output to the /tmp/ directory; you can determine how big it gets by putting a mount point on /tmp/ for sizing. Pagent puts out some status messages every so often; z/OSMF (IIRC) put out a ton of start up messages; omproute and CFZCIM and snmpagent all put messages into syslogd but I don't think they typically put out a lot of messages. A 10-megabyte /tmp/ works for us; we IPLed our production LPAR on June 17th and a month (almost) later syslogd is 1.7 megabytes so we can run about 5 months or so without issues; I don't recall if we've set up archiving in the syslog.conf file or not. |
|
| Back to top |
|
|
|
|
