If the facts do not match your view – do not ignore the facts.

It is easy to ignore facts if you do not like them, sometimes this can have major consequences.    I recently heard two similar experiences of this.

We were driving along in dense fog trying to visit someone who lived out in the country.  We had been there in the daylight and thought we knew the way.  The conversation went a bit like the following

  • It is along here on the right somewhere – there should be a big gate
  • There’s a gate.  Oh they must have painted it white since last time we were here
  • The track is a bit rough, I thought it was better than this.
  • Ah here’s another gate.   They must have installed it since we were here last.
  • Round the corner and here we – oh where’s the house gone?

Of course we we had taken the wrong side road.  We had noticed that the facts didn’t match our picture and so we changed the facts. Instead of thinking “that gate is the wrong colour” we thought “they must have painted the gate”.  Instead of “we were not expecting that gate” we thought “they must have installed a new gate”.  It was “interesting” backing the car up the track to the main road in the dark.

I was trying to install a product and having problems.  I had already experienced a few problems where the messages were a bit vague.  I had another message which implied I had mis-specified something.  I checked the 6 characters in a file and thought “The data is correct, the message must be wrong, I’ll ignore it”.   I gave up for the day.  Next day I looked at the problem, and found I had been editing the wrong file.  The message was correct and I had wasted 3 hours.

Do not restart with the fire hose set to maximum.

There is an article in The Register, about an outage at the Tokyo stock exchange.  One of the problems was that they did not have a process for restarting the environment.  The impact of restarting a system is often overlooked, and in the panic of “get it started as quickly as possible” things can go wrong.   The fire brigade slowly increases the pressure in a fire hose to stop the fire crew from being knocked down with the sudden flow.

TCP/IP is good because it has a “slow start” protocol.  Once a connection has been established, and is working well, the exchange can use bigger buffers, and send more buffers before waiting for the acknowledgement.   This boosts the throughput.  If the back-end is slow to process the data, TCP  slows down the traffic, and then increases the throughput again if the connection can handle it.  If the connection stops and restarts, the rate starts slowly and builds up, rather than use the rate just before the outage.

You cannot expect WAS/CICS/DB2/MQ/IMS to restart at maximum speed; it has to work up to it.  Transactions may have to warm up. There can be many reasons:

  1. Data many need to be read from page-sets into buffers, for example read hot Db/2 data into memory.
  2. Java code needs to warm up to become more efficient (JITed).
  3. The systems need to establish a working set, for example making a buffer pool larger.
  4. Establishing connections may have some serialisation delays.

Restarting faster than a system can cope can cause a domino effect.  A transaction server is restarted and the fire hose of data is turned on.   The transaction server is still warming up, and cannot cope with the volume of requests.    Work for this system is then routed to another transaction server which could handle the workload if the volume gradually increases,  If it gets this additional work  all at once, this instance slows down, and the work is routed to another transaction server etc.

MQ can be seen as the bad guy here.  When you restart MQ, it can go to fire hose mode immediately.   You should start the output channels first to start draining messages, then gradually start the input channels.   If you start the input channels before the output channels, you may get queues and page sets filling up, before the output channels can process the messages.

If you have a policy that all client connects must disconnect and reconnect a random time between15 minutes and 45 minutes this should help spread the load, and gradually you should get a balanced environment.

Do we still need the wine maker’s nose, the mechanics ear and the performance analyst’s glasses?

When I left University one of my university friends went into the wine industry.  We met up a few years later and said that his nose was more useful than his PhD in Chemistry.  Although they had moved towards gas chromatography (which gave you a profile of all of the chemicals in the wine), this was good at telling you if there were bad chemicals in the brew, but not if it would be a good vintage, for that they needed the human nose.

My father would tune his motorbike by listening to it.  He said the bike would tell you when you had tuned it just right, and got it “in the sweet spot”.  These days you plug the computer in and the computer tells you what to do.  A friend of mine had an expensive part replaced, because the computer said so.  A week later he took the car back to the garage because the computer “knew” there was a problem with the same expensive part, and said it should be replaced.    This time the more experienced mechanic cleaned a sensor and solved the problem.  Computers do not always know best.

When I first started in the performance role, the RMF performance reports were bewildering.   These reports were lots of numbers in a small font (so you needed your glasses).  Worse than that, they had several reports on the same page, and to a novice there was a blur of numbers.  Someone then helped me with comments like, you can ignore all the data, except for this number  3 inches in and 4 inches down.  That should be less than 95%.  On this other page – check this column is zeros, and so on.  As you gain more experience in performance, you get to know the “smell” of the data.  It just needs a quick sniff test to check things are OK.  If not, then it takes more time to dig into the data.

There are many tools for processing the SMF data and printing out reports full of numbers, but they add little value.    “The disconnect time is 140 microseconds” – is this good, or is it bad, it better than a disconnect time of 100?.    If the tools were smart enough to say “The disconnect time is 140 microseconds. This value should typically be zero” then this give you useful information instead of just data.

If you think that they could control the Starship Enterprise from one operations desk, they clearly did not have all of the raw data displayed.  It must have been smart enough to report “The impulse engines are running hot: colour red, suggest you reduce power”, because that is what Scottie the engineer kept saying.

If there were smart reports of the problems rather than just displaying data, it would reduce the skill needed to interpret reports, and the need for the performance analyst’s glasses.  Producing these smart reports is difficult and needs experience to know what is useful, and what is just confusing.

Sometimes it feels like the statistics produced have not been thought through.  One example I recently experienced; there is a counter of  the number of reads+writes to disk rather than cache.  For reads, there should be no reads from disk.  For writes, it may be good to write directly to disk, and not flood the cache.  Instead of one number for reads, and one number for write, there is one number for both.  So If I had 10 disk reads, 10 disk writes and 10 disk accesses – is this good or bad ? I don’t know.   This is not a head banging problem, as you usually have only reads or only writes – but not both.  I just had to use my nose,  10 million would be a problem, just 10 – not a problem, and I’ll still need my glasses.

 

What data set is my C program using?

I wanted to know what data set my C program was using.  There is a facility BPXWDYN: a text interface to dynamic allocation and dynamic output designed for REXX users, but callable from C.

This is not very well documented, so here is my little C program based on the sample IBM provided.

The documentation says use  RTARG dsname = {45,”rtdsn”};  but this is for alloc.  With “info” it gives the error message IKJ56231I TEXT UNIT X’0056′ CONTAINS INVALID KEY .  Which basically means rtdsn is not value.     I had to use RTARG dsname = {45,”INRTDSN”};

#include <stdio.h> 
#include <stdlib.h> 
#include <string.h> 
#include <errno.h> 
int main(int argc, char * argv[]) { 
  typedef int EXTF(); 
  #pragma linkage(EXTF,OS) 
  EXTF *bpxwdyn=(EXTF *)fetch("BPXWDY2 "); 
  int i,j,rc; 
  typedef struct s_rtarg { 
    short len; 
    char str[260]; 
   } RTARG;
  char *info ="info DD(APF1) "; 
   
  RTARG dsname = {45,"INRTDSN"}; // not  rtdsn as the doc says
  RTARG ddname = {9,"INRTDDN"}; // not rtddn as the doc says
  RTARG volser = {7,"INRTVOL"}; 
  RTARG msg = {3,"MSG "}; 
  RTARG m[4] = {258,"msg.1",258,"msg.2",258,"msg.3",258,"msg.4"};
 
  rc=bpxwdyn(info,&dsname,&ddname,&volser,
      &msg,&m[0],&m[1],&m[2],&m[3]); 
  if (rc!=0) printf("bpxwdyn rc=%X %i\n",rc,rc);

  if (*ddname.str) printf("ddname=%s\n",ddname.str); 
  if (*dsname.str) printf("dsname=%s\n",dsname.str); 
  if (*volser.str) printf("volser=%s\n",volser.str); 
  for (i=0,j=atoi(msg.str);i<j && i<4;i++) 
     printf("%s\n",m[i].str); 
  
  return; 
}

 

“To infinity and beyond” and how to avoid a whoopsie.

I was reminded of how hard it is to predict the capacity needed for a workload when I read the news about the UK government web site which allowed you to enter your post code, and it told you what level of lock down you were in.  It crashed under the workload.  It should have been clear that within minutes of the web site being announced, a few million people would try to use it.  (It might have been better to have a static web page with all the information rather than try to provide a data base lookup).

Someone told me of a US company whose marketing company had a 1 minute commercial in the interval of the US Super-bowl competition.  They told this to the IT department a week before the game!   The audience of the game is about 100 million people.  If 1% of these people click on a web site within 2 minutes of the advert, this 10, 000 hits per second!  The typical web activity for the company was about 50 hits a second.   After the initial cries of disbelief, the IT department with the help of a large multinational IT company got the additional capacity, and hardware to do load balancing, and got through the night.

One bank said they took the average transaction rate and tested to twice this.  There was a discussion about what the average rate means.   Over a period you have highs and lows.  There is a rule of thumb (I don’t know whose thumb) which says on average, the peak is typically 3 times the average.  Within a peak period (for example 1 hour) looking at a second by second, there will be peaks within peaks.  The rule of thumb said you should plan to support 3 * 3 = (10) times the sustained average.

This bank then worked with IBM to replicate the environment within IBM and run a test to see where the bottlenecks and snags were, and ramped up the workload till they met their targets.  I remember looking at the MQ data, and found the same snags in MQ as we had spotted when we looked at their MQ system a couple of years before.  Logs were not stripped, and they had badly tuned buffer pools. 

Another customer’s test system had more capacity than the production system, so they tested weekly at production volumes + 25%. Many customer’s test systems are much smaller than production and operate on the pray system, where they hope and pray they will not have a problem.

How hard is it to delete lots of data sets? – Easy!

I was configuring a product and had some problems, so I needed to clean up.  I had hundreds of VSAM clusters.  I started using ISPF 3.4 and using the delete line command, but you have extra typing to do for VSAM files, so I gave up.

I had a faint memory of using a MASK to delete things, and a quick search gave me

//S1 EXEC PGM=IDCAMS 
//SYSPRINT DD SYSOUT=* 
//SYSIN DD * 
DELETE COLIN.O.RTE.RK* MASK 
/*

Which deleted all my data sets.

Wasn’t this easy!

How do I do things with a subset of PDS members matching a pattern?

There are some clever things you can do on a subset of members of a PDS.

If you use ISPF  (Browse) or ISPF 2 (Edit) you can specify a data set name of

  • ‘COLIN.AAA.PROCLIB(%%%%%%00)’ and it displays only the members ending in 00.
  • ‘COLIN.AAA.PROCLIB(*AH*)’ to display all member with an AH in the name.
  • ‘COLIN.AAA.PROCLIB’  for all of the members.

If you use ISPF 3;4 I havent found a way of doing the same.

Acting on a subset.

If you have a list of members, for example ISPF 1,2,3;4  you can issue a primary command

sel *99 e

which says select all those members ending in  99, and use the command “e” in front.  Similary  sel %%%%%%00 b.

Sorting the list

You can sort the list by many fields, name, size last changed.  For example “Sort Name”.

I have “Tab to point-and-shoot fields” enabled.   I can tab to column headers, and press enter.   The rows are sorted by this column.

I often use “sort changed” to find the ones I changed recently, and “sort id” to see who else has been changing the members.

Srchfor

I use “srchfor ” or “srchfor value” to look for the members containing a string (or two).

When this command has completed tab to “prompt” and press enter, or enter “sort prompt” to sort the members with hit to the top of the list.

Refresh

If the member list has changed, you can use “refresh” to refresh it.

 

 

How do I compare the directories of two PDS(E)s?

I wanted to compare  two directories to find the differences.   I could see that the number of members was different, but it was hard to see what was missing.

I browsed the web, and found that this was a commonly asked question, and often the solution was to write some Rexx and use the ISPF LM* functions.  I felt this was the wrong way.  

I had used Superc to compare members of different files – could it tell me same information about the member list – yes!

SuperC has different compare types

  1. File –  Compares source data sets for differences, but does not show what
    the differences are.
  2. Line – Compares source data sets for line differences.  It is record-oriented and points out inserted or deleted lines.
  3. Word – Compares source data sets for word differences.  If two data sets contain the same words in the same order, SuperC considers them to be identical, even if those words are not
    on the same lines.
  4. Byte – Compares source data sets for byte differences.  This compare type is most useful for comparing  machine readable data.

Example output of the File comparison type.

NEW: COLIN.ZZZ.PROCLIB  OLD: HLQ.Y.ABCNPARU                                                                                     
MEMBER SUMMARY LISTING (FILE COMPARE)                                                                                     
DIFF SAME MEMBERS   N-BYTES O-BYTES N-LINES O-LINES  HASH1 HASH2 
                                                                                     
 **       ABC11111   171120  173200    2139    2165  78D5C 1113D
      **  ABC9999       640     640       8       8  AB58A AB58A

We can see

  • ABC1111 is different because the “**” in the DIFF column, and the hash code at the right is different
  • ABC9999 is the same in each because the “**” is in the SAME column, and the hash value is the same

You also get a summary of differences

   10   TOTAL MEMBER(S) PROCESSED AS A PDS 
    1   TOTAL MEMBER(S) PROCESSED HAD CHANGES 
    9   TOTAL MEMBER(S) PROCESSED HAD NO CHANGES 
    9   TOTAL NEW FILE MEMBER(S) NOT PAIRED 
  179   TOTAL OLD FILE MEMBER(S) NOT PAIRED

List of members not in both

MEMBER SUMMARY LISTING (FILE COMPARE)                                 
NON-PAIRED NEW FILE MEMBERS | NON-PAIRED OLD FILE MEMBERS               
     ABC$$$$$               |       ZAA$$$$ 
     ABCSCLRR               |       ZYZAPST5 
                            |       ZYZAPST6
  • Member ABC$$$$ and one other are in the “new” PDS, but not in the “old” PDS.
  • Member ZAA$$$$ and 2 others are in “old” PDS, but not in the “new” PDS.

Like most things – easy – once you know how do to it!

Using Line mode

When I used line mode I got output like

                                                  N-LN# O-LN# 
I - SYSNAME                      &SYSNAME.        00004 00003
D - SYSNAME                      S0W1

For one member, the  “new-file” at line 4 was similar to the line in the “old-file” at line 3.

To get from the old file to the new file, delete the line with S0W1 in it and insert the line with &SYSNAME.

Avoiding I/O by caching your PDSEs (It might not be worth it)

When you use most PDS datasets, the data has to be read from disk each time. (The exception is data sets in the Linklist LookAside(LLA) which do get cached. This blog post explains the set up to get your PDSEs cached in z/OS. There is a Red book Partitioned Data Set Extended Usage Guide SG24-6106-01 which covers this topic.
One of the benefits of using a PDSE is that you can get the data sets cached in Hiperspace in z/OS memory.

A C program I am working on takes about 8 seconds to compile in batch, and spends less than half a second doing I/O, so caching your PDSEs may not give you much benefit. You should try it youself as mileage may vary.

SMSPDSEs

The caching of information for PDSEs is doing in the SMSPDSE component of SMS.

You can have two addresses spaces for caching PDSE data sets

  1. SMSPDSE caches the directory of PDSE data sets.  It also caches PDSEs that are contained in the LNKLIST.  SMSPDSE is configured using the parmlib concatenation member IGDSMSxx.  If you want to change the configuration you have to re ipl.
  2. SMPPDSE1. This is used to cache other eligible PDSEs.   SMSPDSE1 is configured using the parmlib concatenation member IGDSMSxx. You can issue a command to restart this address space, and pick up any parameter changes – this is why is is known as the restartable address space. 

It is easy to create the SMPDSE1 address space.  It is described here.

Making PDSE data sets eligible for caching.

It is more complex than just setting a switch on a data set.

The Storage Class controls whether a PDSE is eligible for caching.  It is more complex than just setting a simple switch.  The eligibility of caching is controlled by the Direct MilliSecond Response time.  (Which means the Response time in MilliSeconds of Direct (non sequential) requests).  If you use ISMF to display the Storage Classes, one of the fields is the Direct MSR.  The documentation says If the MSR is < 9 then the value is “must cache”, 10 -998 “may cache”, 999 “never cache”.  I only got caching if MSR was <= 9.

If you change the Storage Class remember to use the command setsms scds(SYS1.S0W1.DFSMS.SCDS) to refresh SMS.
Change your data set to use the appropriate Storage Class with the valid Direct MSR.

By default the SMSPDSE1 address space caches the PDSE until the data set is closed.  This means that PDSEs are not cached between jobs.  You can change this using the commands

setsms PDSE1_BUFFER_BEYOND_CLOSE(YES)
VARY SMS,PDSE1,RESTART

Or just update the parameter in the parmlib IGDSMSxx member.
If you now use your PDSE it should be cached in Hiperspace.

You can use the command d sms,pdse1,hspstats to see what is cached.

This gave me

D SMS,PDSE1,HSPSTATS                                                   
IGW048I PDSE HSPSTATS Start of Report(SMSPDSE1) 531                    
HiperSpace Size: 256 MB                                                
LRUTime : 50 Seconds   LRUCycles: 200 Cycles                           
BMF Time interval 300 Seconds                                          
---------data set name-----------------------Cache--Always-DoNot       
                                             Elig---Cache--Cache
CSQ911.SCSQAUTH                                N      N      N         
CSQ911.SCSQMSGE                                N      N      N         
CSQ911.SCSQPNLE                                N      N      N         
CSQ911.SCSQTBLE                                N      N      N         
CBC.SCCNCMP                                    N      N      N         
CEE.SCEERUN2                                   N      N      N
COLIN.JCL                                      Y      Y      N         
COLIN.SCEEH.SYS.H                              Y      Y      N         
COLIN.SCEEH.H                                  Y      Y      N         
PDSE HSPSTATS  End of Report(SMSPDSE1)

The CSQ9* data sets are PDSEs in Link List.  The COLIN.* data sets are my PDSEs in storage class SCAPPL.   They have Always Cache specified.  If you restart the SMSPDSE1 address space, the cache will be cleared.

You can use the commands

  • d sms,pdse1,hspstats,DSN(COLIN.*)  to display a subset of data sets
  • d sms,pdse1,hspstats,STORCLAS(SCAPPL) to display the data sets in a storage class

SMF data on datasets

There were SMF 42.6 records for the SMSPDSE1 address space showing I/O to the PDSEs.
My jobs doing I/O to the PDSEs did not have a record for the PDSE in the SMF 42.6.

SMF data on SMSPDSE* buffer usage

Below is the printout from the SMF 42 subtype 1 records.

  • BMF:==TOTAL==
    • Data pages read: 20304 read by BMF: 567 <not read by BMF: 19737 ( 97 %) >
    • Directory pages read: 649 read by BMF: 642 <not read by BMF: 7 ( 1 %) >
  • SC:SCBASE
    • Data pages read: 183 read by BMF: 0 <not read by BMF: 183 (100 %)>
    • Directory pages read: 64 read by BMF: <60 not read by BMF: 4 ( 6 %) >
  • SC:SCAPPL
    • Data pages read: 567 read by BMF: 567 <not read by BMF: 0 ( 0 %) >
    • Directory pages read: 472 read by BMF: 472 <not read by BMF: 0 ( 0 %) >
  • SC:**NONE**
    • Data pages read: 19554 read by BMF: 0 <not read by BMF: 19554 (100 %)>
    • Directory pages read: 113 read by BMF: 110 <not read by BMF: 3 ( 2 %)>

We can see that for Storage Class SCAPPL all pages requested were in the cache.

Will this speed up my thousands of C compiles ?

Not necessarily.  See the problems I had.

  • The C header files are in a PDS – not a PDSE, so you would have to convert the PDSs to PDSEs
  • The C compiler uses the SEARCH(“CEE.SCEE.H.*”) option which says read from this library.   This may override your JCL if you decide to create new PDSEs for the C header files.
  • When I compiled in USS my defaults had SEARCH(/usr/include/).  This directory was on ZFS.Z24A.VERSION a ZFS file system.   The files on the ZFS may be cached.

When I ran my compile,there were 31 SMF 42.6 records for CEE.SCEE.H, giving a total of 111 I/Os, there were 2 records for CEE.SCEE.SYS.H with a total I/O count of 14.  If each I/O takes 1 millisecond this is 125 milliseconds doing disk I/O to the PDS, so I expect it is not worth converting compiles to use PDSEs and caching them.

 

Why does my C compile fail if I remove data sets I do not use?

I was playing with caching of header file PDSEs when I compiled a C program. I could see from the SMF 42.6 records that CEE.SCEEH.H PDS was being used.  It took nearly two hours before my job did not use this PDS!
I created a PDSE called COLIN.SCEEH.H and copied CEE.SCEEH.H into it.  I updated my JCL to use the new libraries,  reran my job and the SMF records show I was till using CEE.SCEEH.H.  Hmm this was very strange.

I renamed CEE.SCEEH.H to COLIN.CEE.SCEEH.H.  Did it work ?   No – I got compile errors, so I renamed it back again.  Removing the data set clearly does not work.

I then spotted in the compiler listing that I had the default SEARCH(//’CEE.SCEEH.+’).   I added SE(//’COLIN.SCEEH.+’)  and thought Fixed it!  No … still not fixed,  it still used CEE.SCEEH…

I had to use C options NOSEARCH, SE(//’COLIN.SCEEH.+’) .  The first option turns off the SEARCH(//’CEE.SCEEH.+’) , and the second one creates a new one.   After a cup of tea and a biscuit I remembered I had hit this about 20 years ago!