20419107-1: NETQOS REPORTER ANALYZER
| 07/27/2011 16:25 | KAHLIL THOMAS |
Wesley is saying that all of the Harvester addresses he sees are valid.
Anyway, it looks like the changes to "schedule_days" that we made may possibl
y have fixed the problem. Although I still see what look like errors in the
ReporterAnalyzerManagerServiceLog file, Wesley says that he got an e-mail say
ing that one of his scheduled reports ran.
He'll check this tomorrow to confirm whether the other scheduled reports are
running now. |
| 07/27/2011 15:16 | KAHLIL THOMAS |
About the "scheduled_next" question, Karen replied with:
Not really a problem.. just that if they are the 1969 tstamp, that means the
report isn't scheduled. The Manager service will go through and schedule the
reports that need to be scheduled once you fix the schedule_days problem. You
can't schedule them yourself. |
| 07/27/2011 15:07 | KAHLIL THOMAS |
About the harvester question, Karen sent the following:
In the agent_definitons_deleted table, there is a column called harvesteraddr
ess.
Get the harvesteraddress entries
Select distinct inet_ntoa(harvesteraddress) from agent_definitons_deleted;
If any of these are NOT their harvesters, delete those entries.
Delete from agent_definitons_deleted where harvesteraddress = whatever; Resta
rt manager service |
| 07/27/2011 14:57 | KAHLIL THOMAS |
I ran the above commands, and I also ran the following from the
wiki article:
update report_definitions
set schedule_days=1
where schedule_type=4
and (schedule_days < '1' or schedule_days > '31');
Now, the schedule_days are all between 1 and 31. However, when I recycle
the Report Manager service, I still see errors in the logs.
The wiki article also says:
check
select reportid,name,from_unixtime(schedule_next) from report_definitions whe
re schedule_type<>0;
The scheduled_next column should be non-zero/non-unix-epoch.
I ran that command, but most of the "scheduled_next" rows I see show:
1969-12-31 18:00:00
I'll check back with Karen to see how to proceed. |
| 07/27/2011 13:49 | KAHLIL THOMAS |
| Reconnecting via gotoassist |
| 07/27/2011 13:43 | KAHLIL THOMAS |
Karen Phan replied with a link to the following document:
http://130.200.31.69/index.php?title=Scheduled_custom_reports_are_not_getting
_sent_out
This article mentioned that the mysql errors were due to problems
during the migration of custom reports during the upgrade.
In order to fix this, the article gives the following mysql commands:
. Backup the report_definitions table:
mysqldump -u -p reporter report_definitions > d:\report_definitions.bak
. To fix, run this in a command prompt on the RA server
update report_definitions set schedule_days=log(2,schedule_days) where schedu
le_type = 5;
update report_definitions set schedule_days=schedule_days+2 where schedule_ty
pe = 6;
update report_definitions set schedule_days=1 where schedule_days=13 and sche
dule_type = 6;
update report_definitions set schedule_days = schedule_days - 1 where schedul
e_type = 3;
update report_definitions set schedule_next = 0;
3. Restart the NetQoS Reporter Manager Service and you should not see report
scheduler errors anymore.
If you are still seeing errors, run the following:
select reportID, schedule_type, schedule_days from report_definitions where s
chedule_type=4;
If there are any schedule_days that fall outside the range of 1-31, change th
em to be within this range.
I'll call Wesley and run these commands. |
| 07/27/2011 12:40 | KAHLIL THOMAS |
CALL 2-010 FROM CAI TO CLIENT CALLBACK
s |
| 07/27/2011 12:40 | KAHLIL THOMAS |
I e-mailed Development on this:
Problem Statement: The customer recently upgraded to 9.0.161. They noticed t
hat automated e-mails from RA are not working. We turned up logging and chec
ked the ReporterAnalyzerManagerServiceLog file. This appears to show that th
ey are able to connect to mysql for just over a minute, then at 9:43:17 they
start getting errors such as:
9:43:17 -1 - - :
Unable to connect to any of the specified MySQL hosts.
MySql.Data.MySqlClient.MySqlException: Unable to connect to any of the specif
ied MySQL hosts. ---> System.Exception: Exception of type 'System.Exception'
was thrown.
at MySql.Data.MySqlClient.NativeDriver.Open()
...
Once they start getting errors indicating that they are in this failed state,
the mysql connection will never recover until the recycle the Manager Service
I attached the ReporterAnalyzerManagerServiceLog to my e-mail, and
provided a link to the supportcig file. |
| 07/27/2011 09:17 | KAHLIL THOMAS |
Wesley says that last night they changed all of the .ini files to point
to the correct port. However, he says that they are getting some stats
that appear to be questionable. For example, he says that he is getting
some "DSA Down" events for devices that are not down.
Wesley says that they can manually send out e-mails, but automatic
e-mails are not going out. He also says that automatic reports don't
appear to be running.
I'll see if we can set up a supportassist session with Stuart Robertson |
| 07/27/2011 08:55 | KAHLIL THOMAS |
Received e-mail:
All setting are back to 3308. Not sure if watchdog is giving appropriate inf
ormation. I would to see if we can have a Virtual sesstion setup early in th
e am to inquire on the condiition and recommendation on what to do with our a
pplication. The auto email is not working at this point either. |
|
No comments:
Post a Comment