Replication Progess in PD

I just tried this on my SPA, it shows you the "beginning" bit of the job log, which has the batch numbers - this is what you want, right?

Not sure if you can do this while the replication is running, though.

I guess you can - try, it will not hurt...

(all the numbers below have been replaced from original to 12345s) 

= = =

# /opt/pddb/bin/psql -U pddb ca
[provide your DB password here]

ca=# select id, workflowstepid, log from jobstep where jobid=12345;
   id   | workflowstepid |                      log
--------+----------------+------------------------------------------------
 12345 |           54321 |

. [01:23:45]
. *** Start: ReplicationInit ***
.
. [01:23:45] Replicating dataselection 12345
. [01:23:46] URL of remote SPA: https://12.34.56.78/spa/ws
. [01:23:47] The metadatafilter was unchanged in your policy.
. [01:23:48] Forwarding to storagepool: 123
. [01:23:49] Forwarding from storagepool: 321
. [01:23:50] Retrieving mirror for dataselection: 123
. [01:23:51] Forwarding batchNumbers:12345, 12346, 12347
. [01:23:52]
...
...

Hi Aaron,

I've just noticed your post which is similar to an issue i am experiencing. It's difficult trying to work out how far through the job is, did you have any joy with the suggestion below or have you worked out any other methods??

Cheers,

Andy

Hi Andy,

I'd like to take credit for this but I can't, I ended up logging this one with Symantec who were also a little puzzled as how to do it (unless you catch it at the very beginning). This is convoluted but it does work and at least until the next release it's the only way to do it. From a shell prompt on the PD server which contains the selection run this:

/opt/pddb/bin/psql -U pddb mb -c "select batchnr,count(id),sum(size) from ds_raw_<DSID> group by batchnr order by batchnr" > /tmp/ds2batchsize.out

Replace the <DSID> (including the brackets, from memory) with the data selection ID, which can be found by bringing up the data selection and looking on the general tab. This produced a file listing all the batch numbers, obviously if you go to the end of the file the last batch number is your answer. (I'm running 6.5 but I imagine it's work for all version)

Kind regards

Aaron

Hi Aaron,

Thanks for that, much appreciated. It looks to work with our version (6.2) but comes back with an error as follows:

relation "ds_raw_9" does not exist

Im assuming the number i need to input is the data selection ID, as you mentioned which in our case is a single number in the data selection general info?? When i try this with the brackets i get the following error:

 error at or near "<" at character 48

This then points to the < character at the following stage ds_raw_<9>

Any further thoughts??

Thanks,

Andy

Just tried again without the _raw section, so just ds_9 and this seems to work.

Any ideas what exactly the 'sum' column refers to? Assuming this is the size of the  batch run, is in bytes?

Cheers,

Andy

Hi Andy,

I'm not sure about that either but I made the same assumption as you did. The strange thing is that the 'sum' has different values for the first 6 or so then all are 10485760 and somewhere down teh track it changes again

Aaron

.....

....