Network Management

Hi All,

at the start of this week a performed a Update of our IMC installation. Before the update we run IMC 7.3 (E0605P04) with the NTA 7.3 (E0505P02) module. Now we run IMC 7.3 (E0605P04) and the NTA module 7.3 (E0506).

The updates setup run without any error, all IMC process start normally.

But when I try to access the website of my IMC server I got an error. The strange thing is the error message itself “The page isn’t redirecting properly” another browser tells me that there are too many redirects.

Has anybody an idea how to fix that issues?

Hello,

Are you running on a non-English Windows OS? Please check the DMA on your IMC server. On the processes tab, I suspect the "jserver" process has not started (that's the web interface process).

If that's the case, there is a known issue on 7.3 E0703 in regards to this.

Here is the solution:

1. Open Windows CMD as Administrator and navigate to "iMC\client\bin"
2. Run the command: cacls jmx.password /P [admin group name]:F // replace [admin group name] with the non-English equivalent of the Administrators group depending on your OS language (see also https://www.microsoft.com/en-us/language/search )
3. Restart IMC via DMA and the "jserver" process should now start up successfully

Note: If the "cacls" command is not found, you need to make sure that "Windows\system32" is part of your command-line %PATH%, as cacls is a Windows file system permission editing command.

Note 2:  You could also run the same command with "icacls" instead. The end result should be the Administrators group having Full Control over the jmx.password file.

Hope that helps.

Hello Justin,

thank your for your quicke replay.

All the server for IMC are running on a english OS Version.

Here you can see thaht all prosess a started.

IMC Processes

Have your other suggestion?

best regards,

Florian

Hi Florian,

I'm not familiar with any such issue yet. While I can't ask you to post any logfiles here, do have a look in your iMC\client\log\imcforeground.log (the jserver 'web interface log') and see if you can find any errors around the time when you are trying to connect to the web interface that might shed light on what is happening here. If necessary you can also set it to debugging via System Configuration > Log Configuration > jserver process to DEBUG temporarily, which will provide more extensive output in the log.

Have you tried using both HTTP and HTTPS, different browsers and from a different client/directly on the IMC server? I recall seeing an error about 'too many redirects' on a system where I had set HTTPS only, and then tried connecting via HTTP.

Hi Justin,

Thank you for your tips, it's strange because when displaying ACL administrators groups have Full rights...
Unfortunately, jserver start correctly but login webpage still not display...

2019-06-06 15:02:06 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew]  sourceSeq.size() = 1
2019-06-06 15:02:06 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew] =static source type 0 data.start. 
2019-06-06 15:02:06 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew] source type 0 has no data.
2019-06-06 15:02:11 [INFO ] [Thread-27] [com.imc.plat.proxy.https.HttpProxyService$2::run] Working threads count=0
2019-06-06 15:02:12 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew]  sourceSeq.size() = 1
2019-06-06 15:02:12 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew] =static source type 0 data.start. 
2019-06-06 15:02:12 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew] source type 0 has no data.
2019-06-06 15:02:13 [WARN ] [http-nio-8443-exec-8] [com.imc.servlet.filters.UrlAccessController::doFilter] Access URL With No OperatorInfo: /imc/login.jsf?null from 0:0:0:0:0:0:0:1
2019-06-06 15:02:13 [INFO ] [http-nio-8443-exec-8] [com.imc.res.device.dao.ResUsedLicenseCount::getUsedCount] resp.vcfPECount is : 0
2019-06-06 15:02:13 [ERROR] [http-nio-8443-exec-8] [com.imc.common.ex.ErrorPageBean::checkLoginErrorException] ErrorPageBean find error : java.lang.RuntimeException: org.springframework.jdbc.UncategorizedSQLException: Hibernate operation: Cannot open connection; uncategorized SQLException for SQL [???]; SQL state [null]; error code [0]; Cannot create PoolableConnectionFactory (Cannot load connection class because of underlying exception: com.mysql.cj.exceptions.WrongArgumentException: Malformed database URL, failed to parse the connection string near ';useOldAliasMetadataBehavior=true'.); nested exception is org.apache.commons.dbcp.SQLNestedException: Cannot create PoolableConnectionFactory (Cannot load connection class because of underlying exception: com.mysql.cj.exceptions.WrongArgumentException: Malformed database URL, failed to parse the connection string near ';useOldAliasMetadataBehavior=true'.)
	at com.imc.component.beanmethod.BeanMethod.executeBeanMethod(BeanMethod.java:120)

Here is the imcforeground beginning log.
Database is MySQL 5.7.9, i update mysql to 5.7.26 error sitll occurs.

Regards,

Hi Justin,

I found the same error messange in my log. I´m running MySql server version 5.6.44.0 and i normaly use the website over HTTP (8080).

2019-06-06 14:35:18 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew]  sourceSeq.size() = 1
2019-06-06 14:35:18 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew] =static source type 0 data.start. 
2019-06-06 14:35:18 [INFO ] [pool-11-thread-1] [com.imc.fault.browser.func.FaultBoardMgrImpl::updataFaultDataNew] source type 0 has no data.
2019-06-06 14:35:21 [WARN ] [http-nio-8080-exec-10] [com.imc.servlet.filters.UrlAccessController::doFilter] Access URL With No OperatorInfo: /imc/login.jsf?null from 10.151.240.30
2019-06-06 14:35:21 [INFO ] [http-nio-8080-exec-10] [com.imc.res.device.dao.ResUsedLicenseCount::getUsedCount] resp.vcfPECount is : 0
2019-06-06 14:35:21 [ERROR] [http-nio-8080-exec-10] [com.imc.common.ex.ErrorPageBean::checkLoginErrorException] ErrorPageBean find error : java.lang.RuntimeException: org.springframework.jdbc.UncategorizedSQLException: Hibernate operation: Cannot open connection; uncategorized SQLException for SQL [???]; SQL state [null]; error code [0]; Cannot create PoolableConnectionFactory (Cannot load connection class because of underlying exception: com.mysql.cj.exceptions.WrongArgumentException: Malformed database URL, failed to parse the connection string near ';useOldAliasMetadataBehavior=true'.); nested exception is org.apache.commons.dbcp.SQLNestedException: Cannot create PoolableConnectionFactory (Cannot load connection class because of underlying exception: com.mysql.cj.exceptions.WrongArgumentException: Malformed database URL, failed to parse the connection string near ';useOldAliasMetadataBehavior=true'.)
	at com.imc.component.beanmethod.BeanMethod.executeBeanMethod(BeanMethod.java:120)
	at com.imc.component.beanmethod.BeanMethod.encodeEnd(BeanMethod.java:79)

Best regards,

Florian

Hi Florian,

That looks like it could be a new bug related to IMC's use of the MySQL JDBC connector:

failed to parse the connection string near ';useOldAliasMetadataBehavior=true'

The connector is using a newer version since the E0703 switch to OpenJDK and thatcould be the culprit if it's a parameter that's no longer supported by the new connector.

One more thing you can check is the deploy\log\dmalog.txt. Do you see any error like:

Caused by: com.mysql.cj.exceptions.InvalidConnectionAttributeException: The server time zone value 'CEST' is unrecognized or represents more than one time zone. You must configure either the server or JDBC driver (via the serverTimezone configuration property) to use a more specifc time zone value if you want to utilize time zone support.

Just checking if you might be running into an issue matching what was mentioned in the other upgrade issue post here, also related to MySQL JDBC connector.

Otherwise I think you should open a support case to look into this further.

Justin,

Thank you for your response.
I have the same error  in dmalog.txt after E0703 installation. Installer canot connect ot database.

I need to set a default-time-zone settings in MySQL server ("+00:00")

@Florian, please keep up to date if you open a case.

Thank you !!

Best regards,

Guillaume

Hi Justin,

Thank you for the information and the help so far. A new bug wasn´t the latest update was not the answer I was hoping for. 

So far i found not a massage identical to the mentioned one.  But I keep digging for a bit.

So far I think we had not any problems related to a MySQL connectors.

I will keep this thread updated to the latest developments.

Best regards,

Florian

Hello,

Regarding the MySQL Timezone issue, here is a workaround for now:

1) Login to MySQL command line (as root)
2) SET GLOBAL time_zone = '+3:00'; 
(where '+3:00' should be changed to this server's timezone offset) 

See also https://stackoverflow.com/questions/26515700/mysql-jdbc-driver-5-1-33-time-zone-issue

Florian, please feel free to reference the post in your support case. We should have this looked at by Engineering asap.

Hello Florian,

Do you have any news from HPe support about this incident ?

Best regards,

Guillaume,

Hello Guillaume,

It looks like this issue is triggered by the NTA Module. We received a new patch NTA E0506P02 today that I can confirm fixed the web server redirect issue for one customer already. It should be published within the next week or two, but if you need it in the meantime, please open a support case and we'll make it available to you.

Hi all, I followed this thread carefully...but I still have a question open: could this issue hit a scenario in which the HPE IMC 7.3 E0605P06 is deployed on Linux+MariaDB (MySQL) and an update to HPE IMC 7.3 E0703 is attempted (provided that deployed modules are updated first to cope with new E0703) or not? in other terms...trying the update from E0605P06 and E0703 - actually - on Linux is a flawless process or there are issues open?

Hello Parnassus,

To get a short list of known bug we encountered in this post :
- Set a time zone in database (bug with MySQL / MariaDB too maybe?)
- Issue with ACL on a file which block jserver process startup
- Issue with WebServer redirection, even if all processes are started -> resolved in future NTA module version

My env. is Windows / MySQL
MariaDB is like MySQL you should check timezone bug
I don't know if NTA bug apply to a linux OS

Regards,

Hello,

The significant change in E0703 was the porting to OpenJDK 11, which has updated system requirements as a result of OpenJDK and no longer supports Windows Server 2008 or RHEL pre-7.x OS - see also https://wiki.openjdk.java.net/display/Build/Supported+Build+Platforms

Keep in mind that MariaDB is not officially supported for IMC, though it can work (as you mention it's just another implementation of MySQL). Do make sure you have a working dbman backup before attempting any upgrade.

It would probably be better not to upgrade to E0703 if you haven't started already, unless you need the features/fixes. There are two known issues with RHEL and MySQL, and I posted the workarounds/fixes here (and in the other thread), and for one of them you would need the new NTA patch (to be posted).

There may be a new base version coming out relatively soon to address the known issues - but don't quote me on that as I'm not involved in it, and you would need to contact your HPE Sales/Account Management for any details about future versions.

---

@jguse wrote: Keep in mind that MariaDB is not officially supported for IMC, though it can work (as you mention it's just another implementation of MySQL). Do make sure you have a working dbman backup before attempting any upgrade.

Sorry, I redacted...my IMC is actually running on MySQL (not on MariaDB) as per HPE IMC installation guide on Linux.

---

@jguse wrote: It would probably be better not to upgrade to E0703 if you haven't started already, unless you need the features/fixes. There are two known issues with RHEL and MySQL, and I posted the workarounds/fixes here (and in the other thread), and for one of them you would need the new NTA patch (to be posted).

There may be a new base version coming out relatively soon to address the known issues - but don't quote me on that as I'm not involved in it, and you would need to contact your HPE Sales/Account Management for any details about future versions.

---

Oh, that's really good to know.

Luckily I'm not actually forced to update to E0703 for features/fixes so I can really wait the next patch release (for both Base application and NTA) to see these bugs solved before attempting any update.

Thanks, Davide.

Hello Justin,

Do you have any news about the availability of new patch NTA ? 

Best regards,

Guillaume,

Hello Guillaume,

Not yet, but I see it isn't posted yet. I'll send a reminder to the publishing folks.

Hello Guys,

I´m sorry that I get back to you now. I had to sort out some issues with me support packets first.

Yesterday I got that out of the way and opened a case. After a few hours I got a call from a Support Guy I wanted some additional Logs from me. And today I got the Patch 02 form the NTA506 Version. After the successful deployment the IMC Server is working again.

I want to thank your all form your help with me problem.

Best regards,

Florian