Re: High disconnect times and PPRC
Hi, Then you should do CQUERY to the secondary devices to find the Global Copy links Regards, *Gonzalo Cengotita* 2016-08-17 12:50 GMT+02:00 Martyn Jones : > The adapters we are seeing with high response times are 0032 and 0532. > > Turns out these are the Async links to a Global Mirror site at a > considerable distance (we weren't aware of this part of the config). So, > the high PPRC response times were a red herring. > > It is definitely a DS8800 site, and they are using PPRC/GDPS .. > > So, we're still at a loss to explain the extremely high disconnect times > we're occasionally seeing (up to 77 seconds, although at very low activity > rates). > > Thanks for all your input. > > Martyn > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
The adapters we are seeing with high response times are 0032 and 0532. Turns out these are the Async links to a Global Mirror site at a considerable distance (we weren't aware of this part of the config). So, the high PPRC response times were a red herring. It is definitely a DS8800 site, and they are using PPRC/GDPS .. So, we're still at a loss to explain the extremely high disconnect times we're occasionally seeing (up to 77 seconds, although at very low activity rates). Thanks for all your input. Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Hi, DMX11 is the serial number of the disk cabinet. Which adapters are you seeing with the high response time? *Gonzalo Cengotita* 2016-08-16 14:16 GMT+02:00 Martyn Jones : > Our client is experiencing delays due to high disconnect times in a PPRC > environment. > > These problems do not appear to be due to high traffic, nor can I see any > obvious pattern to their occurrences. > > Looking at the PPRC statistics from the RMF type 74-8 records, I can see > two adapters out of the 10 used that have average responses up to 42 times > those of the other links. > > Using the CQUERY command however, I can find no devices that show these > links being used (on any sub-channel). > > Anybody got any suggestions where else I could look? > > Martyn > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Sure. See below... R748LPSO (PPRC Send Operations) and R748LPST (PPRC Send Active Time) are fields in the RMF 74 Subtype 8 records. I guess the DMX11 in the CU Id could be a hint that it might not be IBM kit. Martyn CQUERY FORMATTED LVL 6 VOLUME REPORT ** PPRC REMOTE COPY CQUERY - VOLUME * (PRIMARY) (SECONDARY) * * SSID CCA LSS SSID CCA LSS* *DEVICE LEVEL STATE PATH STATUS SERIAL# SERIAL#* *-- - -- --- -- * * 43C0 . SMPLEX.CAS INACTIVE 4380 C0 0D .. * * SCH(0) ... ... 000DMX11 * * ... ... * * PATHS SAID DEST STATUS: DESCRIPTION * * - - -- --- * * 0 00NO PATH * * 00NO PATH * * 00NO PATH * * 00NO PATH * * 00NO PATH * * 00NO PATH * * 00NO PATH * * 00NO PATH * * SUBSYSTEM WWNN LIC LEVEL * * --- --- * * PRIMARY 5005076306FFD39B7.7.51.48 * * SECONDARY.1 5005076306FFD278* VOLUME REPORT ** PPRC REMOTE COPY CQUERY - VOLUME * (PRIMARY) (SECONDARY) * * SSID CCA LSS SSID CCA LSS* *DEVICE LEVEL STATE PATH STATUS SERIAL# SERIAL#* *-- - -- --- -- * * 43C0 PRIMARY.. DUPLEX.CAS ACTIVE.. 4380 C0 0D 438B C0 0D * * SCH(0) CRIT(NO)... CGRPLB(YES) 000DMX11 000DBR81* * INCRES(NO). AUTORESYNC(NO). * * PATHS PFCA SFCA STATUS: DESCRIPTION * * - - -- --- * * 8 0031 0031 13PATH ESTABLISHED... * * 0102 0102 13PATH ESTABLISHED... * * 0202 0202 13PATH ESTABLISHED... * * 0203 0203 13PATH ESTABLISHED... * * 0331 0331 13PATH ESTABLISHED... * * 0402 0402 13PATH ESTABLISHED... * * 0403 0403 13PATH ESTABLISHED... * * 0531 0531 13PATH ESTABLISHED... * * SUBSYSTEM WWNN LIC LEVEL * * --- --- * * PRIMARY 5005076306FFD39B7.7.51.48 * * SECONDARY.1 5005076306FFD278* -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Hi I work with IBM DS8* and this links (R748LPST/R784LPSO) are not familiar to me. That's why I was asking . May I see a Cquery output? *Gonzalo Cengotita* 2016-08-16 14:16 GMT+02:00 Martyn Jones : > Our client is experiencing delays due to high disconnect times in a PPRC > environment. > > These problems do not appear to be due to high traffic, nor can I see any > obvious pattern to their occurrences. > > Looking at the PPRC statistics from the RMF type 74-8 records, I can see > two adapters out of the 10 used that have average responses up to 42 times > those of the other links. > > Using the CQUERY command however, I can find no devices that show these > links being used (on any sub-channel). > > Anybody got any suggestions where else I could look? > > Martyn > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
That's a very good question. We were told IBM, but I'm beginning to have my doubts. Trying to figure out if I can tell from the z/OS side. Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Hi, what kind of disk cabinet are you using: IBM disks or any other? *Gonzalo Cengotita* 2016-08-16 14:54 GMT+02:00 Martin Packer : > If you're looking at 74-8 you might want to see the RATE to those > adapters, as well as the Response Time. My - admittedly embryonic - code > does that. Near-zero-rate would bear out your other source of information. > > Non-zero rate might point to an issue on the links. In the past I had to > tell a customer their cross-town links not rated to currents speeds were > in fact failing. 1Mbit vs 4, I think. > > But I'll admit I don't know what "CQUERY" is. > > Cheers, Martin > > Martin Packer, > zChampion, Principal Systems Investigator, > Worldwide Cloud & Systems Performance, IBM > > +44-7802-245-584 > > email: martin_pac...@uk.ibm.com > > Twitter / Facebook IDs: MartinPacker > > Blog: > https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker > > Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or > > https://itunes.apple.com/gb/podcast/mainframe-performance- > topics/id1127943573?mt=2 > > > > From: Martyn Jones > To: IBM-MAIN@LISTSERV.UA.EDU > Date: 16/08/2016 13:16 > Subject:High disconnect times and PPRC > Sent by:IBM Mainframe Discussion List > > > > Our client is experiencing delays due to high disconnect times in a PPRC > environment. > > These problems do not appear to be due to high traffic, nor can I see any > obvious pattern to their occurrences. > > Looking at the PPRC statistics from the RMF type 74-8 records, I can see > two adapters out of the 10 used that have average responses up to 42 times > those of the other links. > > Using the CQUERY command however, I can find no devices that show these > links being used (on any sub-channel). > > Anybody got any suggestions where else I could look? > > Martyn > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > > > > Unless stated otherwise above: > IBM United Kingdom Limited - Registered in England and Wales with number > 741598. > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Martin I have no idea, and no visibility beyond the mainframe systems. That's a question that would have to be asked of the clients network and/or storage guys - who're all on holiday. Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Right. But is the fabric rated the same (or higher)? Cheers, Martin Martin Packer, zChampion, Principal Systems Investigator, Worldwide Cloud & Systems Performance, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 From: Martyn Jones To: IBM-MAIN@LISTSERV.UA.EDU Date: 16/08/2016 15:23 Subject: Re: High disconnect times and PPRC Sent by:IBM Mainframe Discussion List Martin They're 8GB/s links. Busiest I can see is 2.5% (ish), from the 74-8s Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Yes. Links showing high responses (R748LPST/R784LPSO) do NOT appear in any of the response to the CQUERY commands. Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Martin They're 8GB/s links. Busiest I can see is 2.5% (ish), from the 74-8s Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Hi, did you executed CQUERY command against 1 device of every LCU in your system? Regards, Gonzalo Cengotita *Gonzalo Cengotita* 2016-08-16 14:16 GMT+02:00 Martyn Jones : > Our client is experiencing delays due to high disconnect times in a PPRC > environment. > > These problems do not appear to be due to high traffic, nor can I see any > obvious pattern to their occurrences. > > Looking at the PPRC statistics from the RMF type 74-8 records, I can see > two adapters out of the 10 used that have average responses up to 42 times > those of the other links. > > Using the CQUERY command however, I can find no devices that show these > links being used (on any sub-channel). > > Anybody got any suggestions where else I could look? > > Martyn > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
Martin Packer wrote: >But I'll admit I don't know what "CQUERY" is. Look in this book 'z/OS DFSMS Advanced Copy Services' this https://www.ibm.com/support/knowledgecenter/SSLTBW_1.13.0/com.ibm.zos.r13.antg000/gr370.htm Groete / Greetings Elardus Engelbrecht -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: High disconnect times and PPRC
If you're looking at 74-8 you might want to see the RATE to those adapters, as well as the Response Time. My - admittedly embryonic - code does that. Near-zero-rate would bear out your other source of information. Non-zero rate might point to an issue on the links. In the past I had to tell a customer their cross-town links not rated to currents speeds were in fact failing. 1Mbit vs 4, I think. But I'll admit I don't know what "CQUERY" is. Cheers, Martin Martin Packer, zChampion, Principal Systems Investigator, Worldwide Cloud & Systems Performance, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 From: Martyn Jones To: IBM-MAIN@LISTSERV.UA.EDU Date: 16/08/2016 13:16 Subject:High disconnect times and PPRC Sent by:IBM Mainframe Discussion List Our client is experiencing delays due to high disconnect times in a PPRC environment. These problems do not appear to be due to high traffic, nor can I see any obvious pattern to their occurrences. Looking at the PPRC statistics from the RMF type 74-8 records, I can see two adapters out of the 10 used that have average responses up to 42 times those of the other links. Using the CQUERY command however, I can find no devices that show these links being used (on any sub-channel). Anybody got any suggestions where else I could look? Martyn -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN