7 | | It is possible to give feedback or contribute enhancements by editing the page. |
| 7 | It is also possible to give feedback or contribute enhancements by editing the page. |
| 8 | |
| 9 | ---- |
| 10 | |
| 11 | == Collect output of smartctl == |
| 12 | |
| 13 | === HElog_smart.pl by H.Ellenberger === |
| 14 | |
| 15 | Here the description from the script author: |
| 16 | |
| 17 | In order to make more valid judgments about modern big disks above 1TB it |
| 18 | becomes more and more important to keep long term data of some parameters like |
| 19 | soft read errors, spinup time etc. |
| 20 | |
| 21 | If a file name is automatically generated using Disk model and serial no., |
| 22 | smart data can easily be accumulated by appending it for each disk to it's |
| 23 | individual file with a separator holding also a time stamp. |
| 24 | |
| 25 | The attached program first scans /dev/disk/by-id/ for scsi/sata drives and for |
| 26 | each drive then executes smartctl -H. If bad health status is reported it will |
| 27 | be appended to the file. |
| 28 | |
| 29 | Then smartctl -a data is gathered and also appended. |
| 30 | |
| 31 | If the data is written to /var/log/SMART/ it could also become zipped and log- |
| 32 | rotated if appropriate config files are added. |
| 33 | |
| 34 | On desktop systems I run that program from /etc/init.d/ after booting and |
| 35 | before shutdown. On uninterrupted systems I recommend adding a cron job |
| 36 | running at least once per day. |
| 37 | |
| 38 | Since I'm not an expert in Perl programming any feedback and enhancements are |
| 39 | welcome. |