Is LSI MegaRAID Management still on last LAN port?

Printable View

06-08-2009, 08:42 PM
silicon

Is LSI MegaRAID Management still on last LAN port?

We use Intel/LSI MegaRAID cards and have been generally happy only being able to use only the last LAN/eth port for remote RAID manegament by Intel's RWC (RAID Web Concole) or LSI's MSM (MegaRAID Storage Manager).

We have just tried DSS 3511 demo and cannot connect RWC/MSM to any LAN/eth port. I know DSS has better GUI support for MegaRAID management these days, but it doesn't do what RWC/MSM can do, so we still need it. Anyone know of any changes before I spend time trying to figure out what's up? Cheers.
06-14-2009, 08:49 PM
silicon

Just downloading new v6 beta & it's readme states that the feature is not available in 64-bit mode - so I guess this is a 64-bit issue. Do we have any idea when there will be 64-bit LSI Web Console compatability? Cheers
06-15-2009, 02:25 AM
To-M

We hope to have this in Sept. release.
08-04-2009, 12:25 AM
silicon

Testing with v6 update4, RWC access on last eth port now seems to present. The "last port" issue still seems to be there, tho whereas previously RWC would detect nothing on the other ports, our quick testing of v6 instead shows an inability to logon, so something's changed.

Back up the suggestion to change to 32-bit mode instead, this has an interesting effect for DSS v6 Lite, that doesn't seem to allow 32-bit mode to be selected/saved at all!
08-04-2009, 12:35 AM
To-M

Can you send in the logs to suppor@open-e.com I want them to see this so we can check the port or access information. I dont think it is on your end as I have seen 2 other cases like this and need to get them to compile the results.

DSS Lite has only 64bit mode, part of the free version specs.
08-04-2009, 09:41 PM
silicon

Thanks.

Quote:

DSS Lite has only 64bit mode, part of the free version specs

. Aha, that explains it then!

Sorry if this is in the wrong forum, but it is LSI/hardware related. Just did more "status monitoring" tests on v6 Lite, and it seems that there is no hardware/Raid agent "listening" during my test. On our standard Intel SSR212MC2 (McKay Creek/"MC2") platform, when popping a PSU, a fan, or a HDD that's part of a Raid set, no events were logged at all. Hardware platform is set correctly. In addition, Status/Hardware/HardwareMonitoring is not displaying any data any more, just waiting. Might just leave v6 Lite for others to struggle with.
...
Just repeated this with a fresh v6 Demo, and again, no logging or notification of major hardware events (as per above). I am not sure the Hardware Monitoring section of the page is working correctly, as it initially displays the usual MC2 information ok, but then stays in a permanent state of refresh/reload (little circle goes round and round, waiting for something?), rather than periodic refreshes like in v5.
08-04-2009, 10:39 PM
silicon

After another v6 Demo test, RWC access is definitely still on the last eth port only. That logon issue I note in the previous post didn't repeat (RWC often does that sort of thing, makes life confusing!), this time it found nothing at all. Cheers.
08-05-2009, 10:37 PM
To-M

Just tested with V6 build 3534 (though I think you are using the latest if you just downloaded it) and the SSR212MC2 and go the following. I did this with Firefox but IE should be the same.

BatteryType iBBU
Voltage 5470 mV
Current 350 mA
Temperature 31 C
Firmware Status 00000028
Fully Discharged No
Fully Charged No
Discharging No
Initialized Yes
Remaining Time Alarm No
Remaining Capacity Alarm No
Discharge Terminated No
Over Temperature No
Charging Terminated No
Over Charged No
Relative State of Charge 40 %
Charger System State 4113
Charger System Ctrl 13
Charging current 350 mA
Absolute state of charge 38 %
Max Error 25 %
Relative State of Charge 40 %
Absolute State of charge 38 %
Remaining Capacity 335 mAh
Full Charge Capacity 831 mAh
Run time to empty 65535 Min
Average time to empty 65535 Min
Average Time to full 86 Min
Cycle Count 12
Max Error 25 %
Remaining Capacity Alarm 100 mAh
Remaining Time Alarm 10 Min
Auto Learn Period 2592000 Sec
Next Learn time 0 Sec
Learn Delay Interval 0 Hours
Auto-Learn Mode Enabled
Date of Manufacture 08/14, 2006
Design Capacity 880 mAh
Design Voltage 4800 mV
Specification Info 33
Serial Number 447
Pack Stat Configuration 0x68a0
Manufacture Name lsia11000a
Device Name *5806-4
Device Chemistry NiMH
IPMI Sensors:
BB +1.2V Vtt 1.2 V
BB +1.5V AUX 1.49 V
BB +1.5V 1.48 V
BB +1.8V 1.8 V
BB +3.3V 3.37 V
BB +3.3V STB 3.3 V
BB +1.5V ESB 1.49 V
BB +5V 5.1 V
BB +12V AUX 12.03 V
BB 0.9V 0.9 V
Baseboard Temp 33 °C / 91.4 °F
Mem Therm Margin 0 °C / 32 °F
P1 Therm Margin -45 °C / -49 °F
Proc 1 Vcc 1.22 V
08-05-2009, 10:40 PM
To-M

Quote:

Originally Posted by silicon

After another v6 Demo test, RWC access is definitely still on the last eth port only. That logon issue I note in the previous post didn't repeat (RWC often does that sort of thing, makes life confusing!), this time it found nothing at all. Cheers.

I talked with engineers on found out that we have support for this for the LSI 8888 ELP, I am asking them to please fix this and note it in the release notes as well.

I heard from someone the Java version will work but RWC has or uses the JRE. So trying to figure that out. Sorry for the bump on this one but trying to get our guys to work on this ASAP.
08-05-2009, 11:09 PM
silicon

Thanks for the update.
After yet another v6 install (3530), the hardware monitoring is working better, but I am sure there is still work to do. Eg. it is not reporting any Disk Status data at the moment (the MC2 uses the Intel SRCSASJV / LSI8888ELP card). In previous versions of DSS it would simply report them as all "empty", I think because the agent expected a SRCSAS144e card and not the SRCSASJV! I also think that if the agent doesn't like the hardware sensor data, it can make DSS GUI unstable. I will try yet again with latest version soon - but maybe it is not really ready for production yet. Cheers.
08-05-2009, 11:14 PM
To-M

When you are done can you send the logs to support and this post link so the engineers can dig into it. Also latest is 3535.
08-06-2009, 02:08 AM
silicon

Done, with latest version.
Just before grabbing the logs, I also repeated the test of pulling: a PSU (there are two), a fan, and a Hdd in RAID set. The server status LED lit yellow and enclosure / RAID card sounded alarms correctly, then restorded to normal, but no events appear to be logged. There's also a screenshot of the HW Monitoring page, eternally waiting for something.

I will repeat this exercise with v5 (eg. 3518) to compare soon, as we have a system ready to ship to a customer now, and v6 would not be a good idea yet.
Cheers.
08-06-2009, 02:15 AM
To-M

Thanks for the logs we got them - please provide the update to the engineers so they can get this back to you. I need them to test this and look at the scripts.
08-12-2009, 09:23 AM
silicon

Another test round using 3518 in 64bit mode (& this time started everything from scratch - no existing vg or config/setup), and replicated the last-port issue, & logs sent to support@open-e.com.
Summary:
- RWC detected MegaRaid agent/device on eth0 & eth1 ok, but login fails on eth1
- Raid vd degradation status changes were correctly logged and notifications sent this time

All in all, a good result considering the previous confusing results, hopping betwewn versions 3278, 3518, & v6updt4, and trying both 32 & 64 bit versions of each. Cheers.
08-13-2009, 12:27 AM
To-M

Thanks for the tests Silicon, what was the ticket # that you received.
08-13-2009, 12:48 AM
silicon

Not sure of the case number sorry, I generally just shift-delete all notifications! (but I do save all the replies).
The title of my email was: "Forum issue re MegaRaid & RWC - logon failure on all bust last eth port. For To-M", hopefully you can search on this. Cheers.
08-13-2009, 01:12 AM
To-M

Found it - 1010890
08-14-2009, 05:09 PM
To-M

Silicon - Just keeping you updated, developers are working on this issue but will need more time. They sometimes notice that it works but only after RWC is running over 20 minutes until it is possible to access it over the NIC and even then the login works randomly or disks are not listed.