TCPware FAQ: General
How can I tell what version of TCPware I'm running?
To determine what version of TCPware you are running, type this command:
$ NETCU SHOW VER
or if you also want to see if any patches have been applied, use this command:
$ NETCU SHOW VER/ALL
How can I find out what patches are available for TCPware?
All released patches for TCPware are available at vms.process.com/eco.html. Leave the fields blank and click submit to see all patches for all versions.
We are planning to upgrade OpenVMS on our system. Does TCPware have to be upgraded as well?
After any major upgrade of OpenVMS, TCPware must be reinstalled after or it will not start. Make sure that TCPWARE:TCPWARE_LOGICALS.COM has been executed and install TCPware with VMSINSTAL as instructed in the Installation and Configuration Guide. After the installation all that is required is a reboot of the system. After the reboot TCPware will startup normally as previously configured and will function as it did before the upgrade.
If I should find out that my system crashed with TCPware drivers found in the registers, what does Process Software need in order to analyze this crash as quickly as possible?
Process Software will need the following information in order to do a speedy crash analysis:
- The VMS version and machine architecture (VAX or AXP)
- The TCPware version that was running and any patches that had been applied, especially any driver patches. The dump cannot be analyzed without this information.
- Has anything unusual occurred recently or is the crash occurring on installing TCPware? Unusual events include:
- Changed hardware configuration on local node or network
- Changed software configuration including:
- TCPware configuration,
VMS SYSGEN,- new applications on server or client
- an unusual load
- Some SDA information could be useful:
$ ANALYZE/CRASH_DUMP SDA> read/exec (don't need this output) SDA> show crash SDA> format @r3 (optional) SDA> format @r5 (optional) - if this shows up as UCB$... structure, then do SDA> show device/address=@r5 SDA> show stack
This information should be sent to Process Software's Technical Support Department to expedite the crash analysis. In most cases a valid system dump from the crash will also have to be provided.
Why am I not able to access the TCPware system via R-Services (RSH, RCP, Rlogin) when the TCPWARE: HOSTS.EQUIV file looks correct as does the user's .RHOSTS file?
Be sure that the TCPware system can obtain a reverse DNS lookup successfully on the IP address of the client system. For example, if you are trying to issue an RSH command from a system with an IP address of 192.168.1.1 to a TCPware server, be sure the command NETCU SHOW HOST 192.168.1.1 returns the hostname that you have in the HOSTS.EQUIV/.RHOSTS files.
What does the following error message mean when my users try to use PING?
ping: socket: permission denied
This error message indicates that the user does not have the necessary privileges to PING. The PING utility uses the ICMP echo and echo reply messages. To use the PING utility, you need BYPASS or SYSPRIV privilege.
In a cluster with multiple system disks does TCPware need to be installed on each system disk?
Yes, TCPware needs to be installed on each system disk due to the fact that the TCPware installation creates files in some of the system directories (e.g., SYS$LIBRARY, SYS$SYSTEM) You can have nodes of the same architecture share the TCPware common files by specifying a common directory when asked during the installation where to install the TCPware common files.
Commands don't work after re-boot
This is because TCPware's commands are set up as symbols by the file TCPWARE:TCPWARE_COMMANDS.COM and this procedure has not been executed by the current process. We suggest that this be added to the system wide login command procedure (by default sys$manager:sylogin.com).