Hal PING is an "application" which uses IP packets in order to transport ICMP Echo and Echo reply packets. It has the same reliability as IP itself - that is, by no means as solid as TCP.
PING is ***not*** handled by the adapter. There ***does*** need to be an active operating system - ***and*** - active IP software such as the IP component of z/OS Communications Server. You may have in mind the handling of 802.2 or NetBIOS TEST frames. The commands which are set up to support sending and receiving the TEST frame get labelled as some sort of "PING" because they behave a bit like the ICMP PING. OS2PING for example was/is one such command for the OS/2 environment. Your suggestions regarding examining the network have some merit - a pun discovered only after the words had been conceived I hasten to say! However, as I suggested, OSA statistics and ICMP statistics also need to be examined to try to compose a more complete picture. Chris Mason On Thu, 13 May 2010 15:33:47 -0500, Hal Merritt <hmerr...@jackhenry.com> wrote: >IIRC, PING is answered by the adapter. There does not need to be an active operating system, so that points to the network. What does a TRACERTE from the host to each server show? Are all the paths (hops) the same? How about a TRACERTE from each server to the host? Do the failure occur at specific times, such as when a large FTP is in progress? > >Intermittent failures may suggest exhaustion of a key network resource. Have your network folks look at dropped packet counts. Don't know if it applies in this scenario, but network folks love to constrain ports. Could be you are running out. > >Is there a server in the path? > > > > > > >-----Original Message----- >From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Sabo, Frank >Sent: Thursday, May 13, 2010 2:49 PM >To: IBM-MAIN@bama.ua.edu >Subject: FTP and systems PING question. > >FTP and systems PING question. > >Environment: > >We have Z9 J01 with one processor with three LPARS, production, test and tech, all three shares one OSA adapter. The systems are running z/OS 1.9. >Looking at my monitor's, the production lpar is only 29.5% busy while the test lpar is 2.2% the tech lpar does not come into play since it my sand box. IO and paging are not very heavy at the time. > >I talked with network personnel and the see not issues on the network at the time failure. > >Problem: > >We have six UNIX boxes that all issue a ping command to the mainframe at the same time before starting the FTP process. Five pings are directed to the production LPAR and one box is directed to the test LPAR. The all five ping's failed pointing to the production side but the test completed successfully. I don't see any TCPIP errors on either LPAR. I can not explain the difference between the two LPARS why one works and the other one fails. > >We currently do this process about 1500 times a day if not more. The process works fine most of the time but every once in a while it fails. >Not sure where to look for the problem the monitors and logs don't show anything that I can see. > >Any help would be appreciated. > > >Frank W Sabo Jr. >SR. Systems Administrator >Giant Eagle Inc. >Phone: 412 967-3764 >Fax: 412 967-6120 >Email: frank.s...@gianteagle.com ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
