When registering the Jan 2016 arc file in Data archive it goes to Nov 2016?
Why does this happens or does this happens?
It would probably help if you were able to include some screenshots of what you are seeing. That said, my best guess is that the confusion is coming from the archive file name containing "Jan 2016", but when you register the archive the start time or end time column shows as Nov 2016.
Note that the name of the archive file, and the actual time span covered by the archive are decoupled from each other. The actual start/end times for the archive file are stored within the header of the file. When an archive file is created on modern versions of PI Data Archive, the archive file is named according to the start time. However, there are certain situations, though not super common, (overwriting archives, reprocessing, etc) where the actual time range of the archive can change, but the name of the file isn't automatically updated. I suspect that this is what happened in your case.
A good rule of thumb is to always verify the start/end times of the archive files with PI SMT (Operation -> Archives) if the archive is registered. If the archive is not registered you can still view the start/end time of the archive offline using pidiag -ahd <path to archive file>
Hope this helps,
Thank you for you reply, as you can see the name of the arc file was for january which is piarch_2016-01-10 to 20 is on november arc files. Did this overwrites? Thank you. format is yyyy-mm-dd hh-mm-ss if I am not mistaken.
From the screenshots, I could see that both the archive files are different with different Start Time, End Time and and Time Duration, archive size etc.... Only thing is that the archive files in the month of November are saved with an older date and time in their name.
Well isn't the name of the archive suppose to be the date on when it archive? correct me if I'm wrong, so the archive name 2016-01-10 on november archive is not a january archive? by which it's november archive but not named its date as november?
I notice that also that both archives are different and time etc are not the same but the name convention is what confuses us since its a auto generated by its date of archive right?
The Start Date-time and End Date-time of the archive represents the time rage for which it stores/ archives process data. When you query for data for a particular time range, PI Data archives looks for those archive files that covers the queried time range with archive start date-time and end date-time and retrieves data.
Based on the naming pattern given in the archive file naming parameters, the archive file will be automatically given a name at the time of its creation.
In your case, I feel that there might be a mistake in saving the archive files, in which a wrong file name was given while manually copying the archive files.
there is also a possibility that the empty archives were created in the month of January and the data was backfilled later at the end of the year. In this case the archives will have created date time (January in this case) and the data in those archives will be based on the start and end date-time range of the archive ( November here).
Well that might be the case, since I don't know how the client achieve that. Their basing the archives on their name and since it is named January they thought it is for January which is not I guess?
But my point is does this overwrites? If the November archives are already registered and they register the archives for January which is not and it goes to November is this critical?
I also check their backup history which they only use NUMARCH/CUTOFF.
Thank you for your help.
I don’t see any possibility of data getting overwritten.
Alright good to know thank you.
Last Modified Time clearly shows that the data was backfilled in the month of November.
So does it overwrites with the same timestamp or not?
Retrieving data ...