SAN Client service takes a long time to start
Greetings,
We set up SAN client for the first time yesterday and it seems to work pretty good. SLES 11.3 master - 7.6.0.1, 5230 Appliance FT Media Server - 7.6.0.1, SLES 11.3 client - 7.5.0.5. However, I find that it takes up to an hour from the time I start the client service until the client is able to use FT. When I run /usr/openv/netbackup/bin/vxlogview -o 200 to look at the logs, I see these messages:
05/13/14 09:25:37.921 [FTClientMgrService::init] Client successfully started (FATClientMgrService.cpp:337)
05/13/14 09:25:37.960 [DiscoveryTaskThread] DiscoveryTask Thread 119613184 Startup
05/13/14 09:25:37.964 [AddDevice] /dev/sg340
Inquiry "SYMANTECFATPIPE 0.1 nbu05"
TargetHBA:LUN:InitiatorHBA = 0:1:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.965 [AddDevice] /dev/sg339
Inquiry "SYMANTECFATPIPE 0.0 nbu05"
TargetHBA:LUN:InitiatorHBA = 0:0:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.966 [AddDevice] /dev/sg383
Inquiry "SYMANTECFATPIPE 1.1 nbu04"
TargetHBA:LUN:InitiatorHBA = 1:1:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.967 [AddDevice] /dev/sg382
Inquiry "SYMANTECFATPIPE 1.0 nbu04"
TargetHBA:LUN:InitiatorHBA = 1:0:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.967 [AddDevice] /dev/sg381
Inquiry "SYMANTECFATPIPE 0.1 nbu04"
TargetHBA:LUN:InitiatorHBA = 0:1:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.968 [AddDevice] /dev/sg380
Inquiry "SYMANTECFATPIPE 0.0 nbu04"
TargetHBA:LUN:InitiatorHBA = 0:0:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.969 [AddDevice] /dev/sg379
Inquiry "SYMANTECFATPIPE 1.1 nbu05"
TargetHBA:LUN:InitiatorHBA = 1:1:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.969 [AddDevice] /dev/sg378
Inquiry "SYMANTECFATPIPE 1.0 nbu05"
TargetHBA:LUN:InitiatorHBA = 1:0:0x50 State = 1 RefCount = 0
05/13/14 09:25:37.983 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 9 retries left
05/13/14 09:25:38.983 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 8 retries left
05/13/14 09:25:39.984 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 7 retries left
05/13/14 09:25:40.984 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 6 retries left
05/13/14 09:25:41.985 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 5 retries left
05/13/14 09:25:42.985 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 4 retries left
05/13/14 09:25:43.986 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 3 retries left
05/13/14 09:25:44.986 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 2 retries left
05/13/14 09:25:45.986 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 1 retries left
05/13/14 09:25:46.987 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg268 0 retries left
.
05/13/14 09:47:25.735 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 7 retries left
05/13/14 09:47:26.736 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 6 retries left
05/13/14 09:47:27.736 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 5 retries left
05/13/14 09:47:28.737 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 4 retries left
05/13/14 09:47:29.737 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 3 retries left
05/13/14 09:47:30.738 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 2 retries left
05/13/14 09:47:31.738 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 1 retries left
05/13/14 09:47:32.738 [DeviceInquiry] called from DiscoverDevices EVPD Page 0x83 "" device name /dev/sg123 0 retries left
And it goes on like this for some time. It appears as though it scans through every /dev/sg device on the system. Is there a way to exclude what it's scanning or reduce the retry attempts? It's only after it's completed this scan that I can use the client over FT. On larger servers with hundreds of SAN devices and multiple paths per device, client startup will take forever.
An nbftclnt bundle IS available for 7.6.0.1 (are you at 7.6.0.1?) but I have no idea if it will help you or not (and nobody's documented it).
Ask your TSE to check out Etrack 3385322.
All those fixes are in 7.6.0.2, by the way, if you want to bypass the EEB route.
(If I should be looking for 7.5.0.5, let me know - I'm trying to multitask and failing on a Friday)