For some non-simple hardware faults, the most effective and fastest way to solve the fault is to collect logs instead of messing around. The common messing methods are 1. reimage the system' 2. Replace the controller; 3. Restart.
This article details how to collect DELL EMC Unity logs and common problems under the graphical interface GUI and command line CLI:
In terms of GUI, it is the simplest graphical interface method:
1. Log in to Unisphere
2. Click System > Service on the left
3. Select Service Tasks and click Collect Service Information
Click the "+" button in the log window to start generating new logs. Then select the corresponding file. Click on the download icon
Click the button to download the file locally
After reaches100%, the system automatically starts downloading the log. Note that you need to select the + number to generate a new full log.
Note: The log file format is: Unity_500_service_data_FNM00160100999_2016-03-31_23_59_59.tar
Unlike VNX and CLARiiON's SPCollects logs, Unity's Service Data includes both Block and File related logs.
related information and some performance log information.
If only one controller is working, the collected log file format is
spb_Unity_300_service_data_CETV3173800017_2023-10-13_13_07_40.tar
Sometimes due to problems with the ECOM module, any graphical interface operation will report an error. At this time, you need to use the command line tool to collect logs. The following is how to collect logs with the command line tool. The login method is no longer introduced. There are two methods, one is to log in with the IPMITOOL tool, and the other is to log in to the command line through ssh. I have introduced the ipmitool method in other documents and will not introduce it here.
The following is the detailed usage of the svc_dc command.
For collecting logs, the easiest thing is to enter svc_dc directly, without worrying about other parameters. The default collected logs are saved in the following location /EMC/backend/service/data_collection
Logs can also be collected in service mode, but this can only collect logs from one controller.
After the logs are collected, you can see the latest collected logs through the graphical interface, and then download them directly to the local computer, or download them through tools such as winscp. For logs in serivce mode, they can only be downloaded through tools such as scp.
Generally speaking, this log is relatively large, usually more than 500M. Once you have the log, you can send it to us, wechat at StorageExpert.
For downtime, unable to start normally, service mode, with logs, you can quickly locate the problem and solve the problem quickly, instead of doing reimage immediately. If the reimage fails, replace the controller. In many cases, the controller cannot start normally due to OS problems. The more messed up the problem is, the more problems will occur and the harder it will be to deal with later.