Patch-ID# 115861-05 Keywords: esm performance san manager 6320 Synopsis: ESM 2.0 SAN Manager management station patch Date: Mar/08/2005 Install Requirements: NA Solaris Release: 8 9 SunOS Release: 5.8 5.9 Unbundled Product: StorEdge Enterprise Storage Management Suite Unbundled Release: 2.0 Xref: Topic: Relevant Architectures: sparc BugId's fixed with this patch: 4896810 4897300 4897444 4897976 4898089 4900203 4900228 4903215 4906749 4914973 4915919 4919393 4920073 4920581 4921276 4922979 4923174 4923176 4924088 4924211 4924487 4924647 4925101 4925550 4925554 4925635 4925674 4926836 4927441 4928893 4930009 4931618 4932953 4933003 4933088 4933279 4934749 4935652 4935689 4942784 4942792 4943548 4947819 4947822 4954289 4956991 4957575 4958302 4958339 4958348 4963017 4963737 4964558 4969975 4970744 4971108 4978157 4982372 4991403 4993606 5015970 5024473 6200780 6212223 Changes incorporated in this version: 6200780 6212223 Patches accumulated and obsoleted by this patch: Patches which conflict with this patch: Patches required with this patch: Obsoleted by: Files included with this patch: /opt/$ESM_BASE/component/tr_mgmt/bin/check /opt/$ESM_BASE/component/tr_mgmt/bin/component_lib.ksh /opt/$ESM_BASE/component/tr_mgmt/bin/install /opt/$ESM_BASE/component/tr_mgmt/bin/postinstall /opt/$ESM_BASE/component/tr_mgmt/bin/runscript.sh /opt/$ESM_BASE/component/tr_mgmt/bin/sstr.cre /opt/$ESM_BASE/component/tr_mgmt/bin/sstr.cre.default /opt/$ESM_BASE/component/tr_mgmt/bin/svrxmlupd.awk /opt/$ESM_BASE/component/tr_mgmt/bin/uninstall /opt/$ESM_BASE/sssm/etc/properties/InstallerService.properties /opt/$ESM_BASE/sssm/etc/properties/device/serack.xml /opt/$ESM_BASE/sssm/etc/properties/device/switch.xml /opt/$ESM_BASE/sssm/lib/esm-topology.jar /opt/$ESM_BASE/sssm/lib/esm.war /opt/$ESM_BASE/sssm/sbin/esmsyslog /opt/$ESM_BASE/sssm/util/cre/components/alarmservice-impl.car /opt/$ESM_BASE/sssm/util/cre/components/asset.jar /opt/$ESM_BASE/sssm/util/cre/components/discovery-impl.jar /opt/$ESM_BASE/sssm/util/cre/components/esm-common.jar /opt/$ESM_BASE/sssm/util/cre/components/esm-jade.car /opt/$ESM_BASE/sssm/util/cre/components/eventps-impl.car /opt/$ESM_BASE/sssm/util/cre/components/hba.jar /opt/$ESM_BASE/sssm/util/cre/components/hds.jar /opt/$ESM_BASE/sssm/util/cre/components/idresolver-impl.car /opt/$ESM_BASE/sssm/util/cre/components/idresolver.car /opt/$ESM_BASE/sssm/util/cre/components/jmstcf-impl.car /opt/$ESM_BASE/sssm/util/cre/components/notification-impl.car /opt/$ESM_BASE/sssm/util/cre/components/oob_util.jar /opt/$ESM_BASE/sssm/util/cre/components/serack.jar /opt/$ESM_BASE/sssm/util/cre/components/services-api.jar /opt/$ESM_BASE/sssm/util/cre/components/simpletcf-impl.car /opt/$ESM_BASE/sssm/util/cre/components/switch.jar /opt/$ESM_BASE/sssm/util/cre/components/t3.jar /opt/$ESM_BASE/sssm/util/cre/population/transient.xml /opt/$ESM_BASE/sssm/util/cre/population/transient.xml.default /opt/$ESM_BASE/sssm/util/pgsql/lib/libalarm.so /opt/$ESM_BASE/sssm/util/pgsql/lib/libalarm.so.1 /opt/$ESM_BASE/sssm/util/pgsql/nsm1/bin/alarmlog.properties /opt/$ESM_BASE/sssm/util/pgsql/nsm1/bin/postgres.env /opt/$ESM_BASE/sssm/util/pgsql/nsm1/ddl/General.ddl /opt/$ESM_BASE/sssm/util/pgsql/nsm1/ddl/LocalizePropsPopulate.ddl /opt/$ESM_BASE/sssm/util/pgsql/nsm1/ddl/createalarmlog.sql /opt/$ESM_BASE/sssm/util/pgsql/nsm1/ddl/dropalarmlog.sql /opt/$ESM_BASE/sssm/util/pgsql/nsm1/ddl/patchadd-01.sql /opt/$ESM_BASE/sssm/util/pgsql/nsm1/ddl/patchrm-01.sql Problem Description: 6200780 6320 upgrade (add second array) does not populate identity table with array01 information 6212223 ESM 2.0 does not raise alarms when a disk in 6320/6920 fails. (from 115861-04) 4978157 Drive details should report dummy if this is the case 4958302 System1 details should use same terminology as actual rack 4982372 SANPilot showing up as storage app instead of switch 4969975 Wrong default location for LnL of HDS Storage Navigator 4993606 Topology Engine Memory Leak 5015970 esm.logic.device.api.FacadeException shortly after system boots up 4991403 Alarm Noise Becomes Unbearable 5024473 tape jobs fail with ESM (from 115861-03) 4954289 CLI: esm asset --location doesn't work 4956991 Link and Launch to the McData management application (EFC Manager) is not implem 4957575 Number of disks in System1 reported incorrectly 4970744 The SRS/SunMc Norifier fails to send topology information to SunMc 4971108 ESM 2.0 needs to modify port to receive notifications from StorADE on System1. 4925635 Add syslog notifier for alarms 4943548 After in-band discovery of 6920, out-of-band scan is not done (from 115861-02) 4947819 Too many errors being printed to storade.log and esm.log 4947822 Topology graph doesn't contain device properties when serialized. 4932953 Description value of some of Event are displayed as English ones 4915919 SnmpNotifier sends bad correlated events 4963017 Performance fixes 4964558 XML parsing performance fix 4958339 Topology page: when I expand a group, all items from action dropdown disappear 4958348 View Snapshot brings up a blank screen. 4963737 Found memory leak in CRE (from 115861-01) 4924211 The input string for LocalizeUtil.getLocalizedString method should be defined i 4898089 localized characters saved in text fields of "Recipient" show as garbage ones 4900228 date-time format on Test Result and Diagnostic pages is a English (US) one 4924088 Event Advanced Filter: There are 2 instances of Diagnosis Start and no D..End 4903215 if esmcontrol starts from zh, in some cases, description text shown as ??? mark 4924647 Event.properties value for "Diagnosis end" is wrong. 4896810 many info messages on esm console are not i18ned 4897300 on the array details, the Actions pull-down menu option "ShowPath" is not i18ned 4897444 the Alarms names on many devices pages are not i18ned 4897976 some table contents on some device pages are not i18n 4906749 Use locale in calls to retrieve data from asset service 4922979 localized sorting in db doesn't work. 4925101 The command esm shows the ???? marks. 4930009 some _en.properties files don't have a corresponding .properties as default one 4925674 esm 2.0 cannot receive events from a solution rack 4927441 T3A fw 1.18/19 can't be instrumented due to vol slice exception 4920581 LnL from SAN Manger to CR of 6910 fails 4921276 Link and Launch to Topology in SAN Manager from a 3910 series array opens with w 4900203 localized characters saved in text fields of asset detail Edit show as ??? marks 4923174 On Test Setup page, options "true" and "false" in the drop-down menus not i18ned 4924487 When Exception thrown on UI, localized characters on UI displayed as ??? marks 4931618 Topology page does not localize alarm state values 4928893 Asset info. shows un-l10ned/garbage contents when client/MS locales differ 4933003 the option from Device Type menu in the Advanced Filter pop-up doesn't work 4920073 many massages displayed on UI depend on the locale smcwebserver 4925554 cannot add switches manually using out-of band discovery 4925550 auto discovery missing 2 out of 3 switches in san 4933279 use 2 or more locales same time will cause mixing language strings in some pages 4923176 In Zones table, the values "Soft" and "Hard" in the Type column are not i18ned 4934749 The value "link" in the Alarm table is not i18ned 4933088 use l10n characters in containing field in the Advanced Filter get no result 4935689 Events: Sorting "Topic" and "Element Type" makes data garble on ja. 4935652 launch SANbox_Manager quits working if you edit (without actually modifying) 4926836 San Manager has "No Contact" with host that has fully confi minnow (1024 lun) 4919393 I18N: CLASSPATH statments for l10n jar files have not added into the sstr.cre 4942784 inband switch data seems to be inconsistently lost 4942792 Launched supporting apps should open in a new browser window 4914973 LnL of Sansurfer from topology page fails. Patch Installation Instructions: -------------------------------- For Solaris 7-8 releases, refer to the man pages for instructions on using 'patchadd' and 'patchrm' scripts provided with Solaris. Any other special or non-generic installation instructions should be described below as special instructions. The following example installs a patch to a standalone machine: example# patchadd /var/spool/patch/115861-05 The following example removes a patch from a standalone system: example# patchrm 115861-05 For additional examples please see the appropriate man pages. Special Install Instructions: ----------------------------- 4926836 San Manager has "No Contact" with host that has fully confi minnow (1024 lun) This bug was caused by a timeout. The remote agent was still gathering data while the management station timed out waiting for it. If you are running into these timeouts, this patch will nott fix the problem, but it makes the timeout value configurable so that it can be fixed. Once you change the timeout to a larger value, the problem should clear up. To see if you are running into this particular timeout problem, look in the /var/opt/SUNWstm/sssm/log/storade.log.* files on the management station for errors that look similar to this: 09/26 15:41.45 com.sun.jade.device.util.DeviceHandler$Instrumentation.isHung:308 ERROR:Device instumentation hung for device: StorAdeHBA_System:809ccxxx If these messages are in the log, then once this patch is applied, you will need to increase the timeout value. Do this by creating a CC_DeviceHandler.hang_time value in the file: /opt/SUNWstm/sssm/etc/properties/InstallerService.properties on the management station. Here is an example of the edit you would make to change the timeouts: CC_DeviceHandler.hang_time=600000 The timeout value is in milliseconds. The example above will increase the timeout to 600000 milliseconds (ten minutes). The default is 300000 (five minutes); this is the value used if a CC_DeviceHandler.hang_time does not exist in the file. The CC_DeviceHandler.hang_time does not exist in the file by default, so it will need to be added if a user wishes to increate the timeout. ----------------------------------------------------------------------------- 5024473 tape jobs fail with ESM A driver conflict could occur if another installed application is accessing the driver stack at the same time ESM is performing an inband query. This patch provides a workaround for this problem, allowing a user to disable the periodic ESM inband queries. There are two inband queries that can occur; here are instructions for disabling both queryies. 1. The first is a periodic discovery for new devices that have been attached to the SAN. To disable this discovery, edit the file: /opt/SUNWstm/sssm/etc/properties/InstallerService.properties on the management station, changing the CC_DeviceHandler.discovery_polling_interval value to a <= 0 value, for example: CC_DeviceHandler.discovery_polling_interval=0 Once this inband query is disabled, ESM will no longer be able to automatically discover newly connected devices to the SAN. New devices may still be discovered through their IP interfaces, however, as follows: To discover newly attached storage and switches: Discover the device by clicking the "Discover" link and entering the IP address of the new device. To discover newly attached hosts: Discover the host by clicking the "Discover" link and entering the IP address of the new host. A one time only inband discovery will be performed on the new host to see what devices are attached. 2. The second inband query that occurs is part of host monitoring. Turning off monitoring for a host will disable its inband queries for device path and HBA information. Monitoring can be turned off and on on the SAN Elements page, or by using the 'esm' command: # esm asset -M [ on | off ] [ -i | -n ] When you turn off monitoring for a host, you will no longer be able to get updated information on HBAs, device paths, or host status. Once you turn monitoring back on, this information will be updated during the next instrumentation cycle. ----------------------------------------------------------------------------- 4925635 Add syslog notifier for alarms This patch makes ESM capable of sending alarms to the syslog. After the patch is installed, this feature must be enabled before any alarms will actually be sent to the syslog. To enable this feature, use the command: /opt/SUNWstm/sssm/sbin/esmsyslog enable To turn off logging to syslog, use the command: /opt/SUNWstm/sssm/sbin/esmsyslog disable The product must be running for the esmsyslog command to run successfully. By default, only alarms of severity Major and above are sent to the syslog. To change the severity threshold, edit the file /opt/SUNWstm/sssm/util/pgsql/nsm1/bin/alarmlog.properties. ESM will need to be restarted for any changes to this file to take effect: /opt/SUNWstm/bin/esmcontrol -m restart Messages are logged as facility 'user' with a level of 'info' as the default. Please see the man pages for syslog.conf(4) for more information about logging facilities and users. README -- Last modified date: Tuesday, March 8, 2005