Reallocated Sector Alerts disabled in RAIDiator 4.2.22?

[jlehtinen]

I upgraded two ReadyNAS 3200's to RAIDiator 4.2.22 about 3 months ago.

I have been using the Status->Logs section of FrontView to monitor Reallocated Sector changes on drives in these units. Since upgrading to 4.2.22, I have not seen any alerts in the logs. I manually checked SMART status on all drives today and Reallocated Sector count HAS increased on several drives with no mention in logs.

This change is misleading and somewhat dangerous, as I was monitoring the health of several drives that I am not confident in. I assumed that because I was not seeing sector count change in the logs that I was OK. You need to inform people when you make changes on how logging is reported in FrontView.

[vandermerwe]

Have you tried 4.2.23 or 4.2.24 to see if this has persisted through these releases?

[jlehtinen]

Not yet, but I guess I'll have to do that. These are storage units in a production environment, so that's no simple task.

Couple points here:

1.) I opened a ticket on this and the tech's response was the same: "Try different firmware versions and see if the alerts start working again". :? Not trying to be a dick, but I'm not interested in doing NetGear's QA. It is disappointing to upgrade firmware to resolve issues and to find out I have to be watching for NEW issues. Also, how do I know I won't introduce even more issues by upgrading?

2.) We've had to replace the chassis on 2 of the 3 units we have in the last year (that's the entire unit minus the drives). I don't know if that's due to actual hardware issues, or if that's the default response when the troubleshooting hits a certain point.

3.) We've seen reoccurring network stability issues that have been unresolved following multiple tickets and many hours of troubleshooting and reconfiguration. Just talked to a tech who mentioned potential issues caused by upgrading firmware without a full volume rebuild. Apparently the older version firmware used a different partition table. Maybe we've got some nasty base level configuration or filesystem issues in these units and that's causing our issues...

Based off my experience with their products I do NOT recommend NetGear storage products.

[mdgm-ntgr]

On the forums there is always going to be a disproportionate number of posts from users having issues rather than from the majority for whom things work fine. No OS is bug free. Also PC hardware can and does fail at any time.

Have you tried changing network cables, changing a port on a switch, even doing a direct-connection.

[jlehtinen]

I understand perfectly well that nothing is perfect and that equipment breaks. My complaints are detailed above and I feel they are clearly worded and really don't need to be repeated.

I'm basing my opinion of NetGear off of my personal experience with the hardware and the company. If you've had a good experience, that's great. I have not.

As for troubleshooting the network issue, please see this thread and respond if you have advice: viewtopic.php?f=118&t=70267

[mdgm-ntgr]

What email provider are you using to send email alerts? Gmail works well for me. The internal provider is blocked by some ISPs.

[jlehtinen]

These units are on a physically isolated gigabit SAN, so I do not use email for alerts.

As original post states, I monitor drive health via FrontView. Reallocated sector alerts used to appear on the Status->Logs page. Now they don't...

[mdgm-ntgr]

If you download the logs (Status > Logs > Download all logs) does the log for cron suggest the daily 4am check for SMART errors has been run?

[jlehtinen]

Sorry for delay here.

Which log file should I check for that info?

I think I'm seeing a weekly check, but not a daily. There's log files that are named : disk_smart_2013_07_28.log, disk_smart_2013_08_04.log, etc.

[mdgm-ntgr]

The script to check the health of the disks is run using cron.

[jlehtinen]

Right...

I have downloaded the logs as you have suggested. Which log file contains the entries you want me to check?

I don't see "syslog" in the log bundle, and none of the logs that have any reference to "cron" in their name.