Re: SMF timestamps

On Wed, 7 Sep 2011 18:26:40 -0400, Robert A. Rosenberg wrote: > >Since each record has a time stamp in its header showing (in local >time) when it was written, I would think that if record X+1's time >stamp is an hour earlier than record X's time stamp that would be a >red flag that the time shi

Re: SMF timestamps

"Ed Gould" wrote in message news:<1315436928.60134.yahoomailmob...@web161421.mail.bf1.yahoo.com>... > Interesting topic. > SMF timestamps are a little bit tricky as they are not when the record is written but when they are created. > So a record could be in storage and n

Re: SMF timestamps

At 09:36 -0500 on 09/06/2011, McKown, John wrote about Re: SMF timestamps: I should have said that I know how to convert the SMFSTAMP8 to RCF3339 format. The problem is determining the timezone offset for a specific date & time rendered in LOCAL time. Now, for my own shop, there is only

Re: SMF timestamps

Interesting topic. SMF timestamps are a little bit tricky as they are not when the record is written but when they are created. So a record could be in storage and not written for as long as a week maybe more. We had one job that lasted a week. We had to change code to allow for this and it

Re: SMF timestamps

I'm using SMF90 subtypes 1 & 2 to catch time and date changes from SET commands. This has worked well so far to provide the logic necessary to identify that the clock has been set back and there will be SMF30s and almost every other record type with an hour duplicated due to Daylight Saving Time

Re: SMF timestamps

I hadn't really planned on sequential processing. That assumes that the relative physical sequence of individual records never change. Which implies not having been run through a sort process. Of course, we have multiple z/OS images and we merge all the SMF data together using DFSORT. This is done

Re: SMF timestamps

Sent: Wednesday, 7 September 2011 4:46 PM To: IBM-MAIN@bama.ua.edu Subject: Re: SMF timestamps Walt Farrell wrote: >The problem with the transition from daylight-saving time to standard time, >Elardus, is that the hour from 1 a.m. to 2 a.m. (0100 to 0200) happens twice: >once while you&

Re: SMF timestamps

Walt Farrell wrote: >The problem with the transition from daylight-saving time to standard time, >Elardus, is that the hour from 1 a.m. to 2 a.m. (0100 to 0200) happens twice: >once while you're on daylight saving time, and once when you've switched back >to standard time. (Assuming I got the t

Re: SMF timestamps

On Tue, 6 Sep 2011 10:47:42 -0500 Paul Gilmartin wrote: > Am I supposed to infer from this discussion that SMF timestamps are > recorded in local time, as opposed to UTC? > > What century is this, anyway? I thought it had long been generally > believed that critical times

Re: SMF timestamps

In <4e663cca.5070...@valley.net>, on 09/06/2011 at 11:31 AM, Gerhard Postpischil said: >The easiest would be to keep existing formats, but add a new SMF >record type Ot new segments in the type 30. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see

Re: SMF timestamps

On Tue, 6 Sep 2011 11:28:40 -0500, Elardus Engelbrecht wrote: > >Damn! While we are not using those daylight saving time thing, I see your >problem... > >What about a table driven solution? Say for hours 1-hours 2 you use W.06, >hours 2 - hours 3 you use W.05? I admit this is *ugly* and may no

Re: SMF timestamps

-6507 On 2011-09-06 09:23, McKown, John wrote: The SMF timestamps in SMF records, such as reader start date& time, where the date is PL4'0cyydddF' and the time is a fullword binary number of 1/100ths of a second past midnight. But what is confounding me is that this seems to be t

Re: SMF timestamps

McKown, John wrote: >I should have said that I know how to convert the SMFSTAMP8 to RCF3339 format. >The problem is determining the timezone offset for a specific date & time >rendered in LOCAL time. Ok. You want to determine what the GMT/UTC is at the time of LOCAL time? If so, sorry that I'

Re: SMF timestamps

On Tue, 6 Sep 2011 11:31:22 -0400, Gerhard Postpischil wrote: > >The easiest would be to keep existing formats, but add a new SMF >record type that would record clock information whenever that is >changed (including the daylight savings transition), and provide >the IPL settings. > Also, include C

Re: SMF timestamps

>Am I supposed to infer from this discussion that SMF timestamps are recorded in local time, as opposed to UTC? It depends on the record/vendor, as Dr. Barry stated. >What century is this, anyway? I thought it had long been generally believed >that critical timestamps should be record

Re: SMF timestamps

27;t have >a job start during the first 02:00-02:59 interval, and another one start at >the exact same time during the second 02:00-02:59 interval. > Am I supposed to infer from this discussion that SMF timestamps are recorded in local time, as opposed to UTC? What century is this, anyw

Re: SMF timestamps

On 9/6/2011 11:04 AM, Richard L Peurifoy wrote: Perhaps we need a requirement for an option to use GMT in the SMF header, or maybe to just use TOD timestamps. It would be nice to have the GMT offset in the records too, but this would require changes in the record structure which will be harder to

Re: SMF timestamps

John, Barry, did a great job of explaining the issue(s). what we did with users is to say all times are where the system (CPU) is. This simplified issues immensely. In our case there were 2 locations one set and in Europe(and. Chicago) Both were fine with the agreement. They were not concerned

Re: SMF timestamps

On 9/6/2011 9:37 AM, McKown, John wrote: I should have said that I know how to convert the SMFSTAMP8 to RCF3339 format. The problem is determining the timezone offset for a specific date& time rendered in LOCAL time. Now, for my own shop, there is only one problem: the 1 hour overlap when conv

Re: SMF timestamps

On Tue, 6 Sep 2011 08:23:12 -0500, McKown, John wrote: >... That is, being in the US Central Timezone, when we go to Daylight Saving >time (from TIMEZONE=W.06 to TIMEZONE=W.05) at 02:00, the time range from >02:00-02:59 repeats. Is this correct? If so, I don't know if 02:10, for >example. is

Re: SMF timestamps

lth Insurance Company.SM > -Original Message- > From: IBM Mainframe Discussion List > [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Elardus Engelbrecht > Sent: Tuesday, September 06, 2011 9:28 AM > To: IBM-MAIN@bama.ua.edu > Subject: Re: SMF timestamps > > McK

Re: SMF timestamps

McKown, John wrote: >The SMF timestamps in SMF records, such as reader start date & time, where the >date is PL4'0cyydddF' and the time is a fullword binary number of 1/100ths of >a second past midnight. But what is confounding me is that this seems to be >the LOCAL

Re: SMF timestamps

There are MANY datetime values in SMF records, and in MANY formats. The 8-byte "SMFSTAMP" format with time to 2 decimals and Julian date is used not only in the standard timestamp in the SMF header bytes 3-10, which is ALWAYS on the LOCAL time zone, but SMFSTAMP8 is used for many other datetime va

SMF timestamps

The SMF timestamps in SMF records, such as reader start date & time, where the date is PL4'0cyydddF' and the time is a fullword binary number of 1/100ths of a second past midnight. But what is confounding me is that this seems to be the LOCAL time, not GMT/UTC (yes, I know GM