I saw the following error in the Infrastructure SystemOut.log each day when the daily or weekly digest is sent on this Connections 5.5 CR2 server.
Weekly Email Digest Report
Tranche ID: tranche_2_5oPldKwZTaR7aAiPFw4L08CyRW
Start Time: 7 Mar 2017 23:00:02 GMT -> End Time: 7 Mar 2017 23:00:02 GMT
Users Processed: 4, Digest Mails Sent: 2
[3/7/17 23:00:04:110 GMT] 00000413 CrudDao E com.ibm.lconn.hpnews.data.dao.impl.ibatis.CrudDao DELETE Batch – doInSqlMapClient Exception caught –
com.ibm.db2.jcc.am.BatchUpdateException: [jcc][t4][3.69.66] Batch failure. The batch was submitted, but at least one exception occurred on an individual member of the batch.
Use getNextException() to retrieve the exceptions for specific batched elements. ERRORCODE=-4229, SQLSTATE=null
I thought that there was a problem with data in the database and being a recent migration I thought there was something “bad” in there.
db2 “select PERSON_ID, DISPLAYNAME, EXID from HOMEPAGE.PERSON where PERSON_ID in (select PERSON_ID FROM HOMEPAGE.EMD_EMAIL_PREFS where TRANCHE_ID = ‘tranche_1_5oPldKwZTaR7aAiPFw4L08CyRW’)”
The above query gave me the users that were in this tranche. I then applied the following trace to see what was being sent
This really didn’t help me too much as the data looked fine. Then I turned to Google and found that Ted Hardenburgh had already come across this problem detailed in IBM Connections 5.5 – Errors in SystemOut.log cleaning up HOMEPAGE database
I raised a PMR and IBM sent me LO90678 which I applied. Checking the SystemOut.log the next day I still saw the same errors. IBM passed this back to development and they responded with, “exceptions should go away after expiration period (default 30 days) for stories. News cleanup service should work normally after this period of time.”
I have checked this morning and the exceptions at 23:00 seem to have stopped.
Internally we originally used a DNS round robin alias for Connections to connect to to route SMTP emails but that was problematic when one of the servers in the alias was taken off line.
IBM has made this easier by allowing you to use MX records to list the SMTP servers to connect to as detailed in Sending mail from any available mail server.
It was fairly simple using the example in the knowledge center to set this up. Firstly I had our network team create (internal only) MX records for three Domino servers for internal.acme.com with the required weightings. Then I checked out notifications-config.xml and edited the following lines and checked it back in.
At first I left the below line in and it didn’t work.
Setting <property name=”mail.debug”>true</property> wrote the following to the SystemOut.log.
[2/21/17 20:13:34:309 GMT] 0000023e SystemOut O DEBUG: getProvider() returning javax.mail.Provider[TRANSPORT,smtp,com.sun.mail.smtp.SMTPTransport,Sun Microsystems, Inc]
[2/21/17 20:13:34:322 GMT] 0000023e SystemOut O DEBUG SMTP: useEhlo true, useAuth false
[2/21/17 20:13:34:322 GMT] 0000023e SystemOut O DEBUG SMTP: trying to connect to host “domino.internal.acme.com.”, port 25, isSSL false
[2/21/17 20:13:34:347 GMT] 0000023e SystemOut O DEBUG SMTP: exception reading response: javax.net.ssl.SSLException: Unrecognized SSL message, plaintext connection?
[2/21/17 20:13:34:348 GMT] 0000023e SystemOut O DEBUG SMTP: useEhlo true, useAuth false
[2/21/17 20:13:34:348 GMT] 0000023e SystemOut O DEBUG SMTP: starting protocol to host “domino.internal.acme.com.”, port 25
[2/21/17 20:13:34:349 GMT] 0000023e SystemOut O DEBUG SMTP: exception reading response: javax.net.ssl.SSLException: Connection has been shutdown: javax.net.ssl.SSLException: Unrecognized SSL message, plaintext connection?
Remming out the aforementioned line referencing allowed me to connect over port 25.
To test this my colleague stopped the SMTP listener on the Domino server with the lowest weighting causing it to connect to the next server.
A customer was asking me about the daily and weekly digest notification emails that are sent out. Their query was going down a different route but to try and head off a potential question about editing the times that the notifications are sent I read around Customizing IBM Lotus Connections 3.0 email digests which was useful but did not give me the information I wanted.
Remembering that I had read something related to this in the last couple of months I went through my feeds and found that one of my ISSC colleagues Dave Hay had already blogged about it in a post IBM Connections 3.0.1 – Email Notifications
“There is no set ‘time’ for the digests, they are processed/delivered throughout the day.
The users in the system are split into 20 tranches.
Each hour, a task runs to pick up the next applicable tranche (i.e. one whose daily digest was not processed in last 24 hours, or whose weekly digest has not run in 168 hours).
You could change the time of the hour that the task runs, but this will not give the behaviour you are looking for.”
Thanks again Dave!