dalelsw

Here is an article that provides a good insight into my comments regarding workload types: An explanation of IOPS and latency (here).

STM is able to publish a report to a local directory (default is the "publish" directory under the webapp ROOT directory, which is not obvious and not documented), or a CIFS share (NFS would also be good). I currently present the publish directly via a locally installed instance of Apache web server. This is only a work…

I would like to see composite views for items such as LUNs. For example, a graph that simultaneously displays several metrics (customizable) for a single chosen LUN. For example (at the same time on the one graph): IOPS, read/write%, latency, and read cache hit%. (Note, this is for reports and graphs, such as when looking…

Will SRM cover fiber channel switches (eg: Brocade and Cisco MDS, etc...)? How well is Storage monitoring integrated with Virtual monitoring for end-to-end analysis (eg: Drive > VM > LUN > Storage Pool > Controller, etc... for performance and capacity, etc...)? Is this SRM related to the EMC SRM product? Counting VMware…

Given that STM is used to monitor/profile some considerable enterprise environments, it would make sense that STM is capable of an equivalent level of High Availability (HA). Some options outside of STM providing the HA itself would be to use a VM in a management cell/cluster with DR fail-over via something like SRM.…

A scheduling mechanism with the flexibility of "cron" would be extremely useful. The current mechanism is close to useless, unless performing daily or monthly one-off reports. Eg:Schedules such as: * Every hour, between 9am to 5pm, on week days. * 30 minutes past every 6th hour (eg: 0:30, 6:30, 12:30, 18:30). * Every 10…

We currently have all polling located on dedicated polling agent servers. While polling servers can still collect data when the STM server is off-line due to database maintenance this is still an issue: 1) We cannot view reports or perform analysis. 2) Alerting for performance related events cannot be escalated (agents can…

The dbutil performs a MySQL analyze and repair/recover on all tables in the storage DB, which results in a DB optimization being done. This is effectively the same from performance/time point of view (dbutil uses "myisamchk", which is the main MySQL tool used to repair, analyze and optimize). Keep an eye on the IOPS…

I see that the SMI-S specification is limited to a minimum period of 15 minutes. However, it would be good if there were a method via the STM GUI to specify lower values. It is possible to unofficially do this by editing ".../agent/systemic/mod.sys.smis.Smis_1.0/mod.sys.smis.Smis.xml" on the agent performing the…

Compare the WWN of the LUN as seen on the server (as a disk), to the WWN of the LUN presented by the array.

Hi, Unless I am missing something, I only stumble across new releases and hot-fixes for STM on the Customer Portal when I log in, or if I happen to see a forum update on Thwack. It would be good if I can opt in via my Customer Portal account to have a "New Release (or Hot-fix) is available for Product X (that you own)"…

The problem with the dbutil is that it keeps STM offline for an extended period of time, especially with a large DB (eg:60GB+). It would be good if the DB optimize could be done while STM is up and on-line. One of the causes here is the use of MyISAM databases. Moving to use InnoDB would help solve some of this as well as…

I would also like to see this. For example, some standardized or abstracted queries. I know that when generating reports we can see the SQL query used to generate the report in the comments of the report HTML page source (eg: show source). However, this still requires a bit a reverse engineering and guess work to find a…

I would also like to see this done in a consistent manner for all monitored storage arrays, and also for other devices such as FC switches (Brocade FCSWs provide component status, but don't alert on the polled information, Cisco MDS FCSWs don't even provide component status, eg: temperature, PSU and fans, etc...).

I agree. Given that STM is used to monitor/profile some considerable enterprise environments, it would make sense that STM is capable of an equivalent level of High Availability (HA). Some options outside of STM providing the HA itself would be to use a VM in a management cell/cluster with DR fail-over via something like…

I see that some collector poll intervals are being raised (eg: SMI-S and VMware). One of the issues we have with the SMI-S collector is that the 15 minute interval is already too big. If we are using STM to alert on performance spikes, we would ideally like it to be 5 minutes or even lower. Also, is there a reason for…

Will STM 5.7 support NetApp 8.1+ Cluster Mode? If not, has this been considered for the STM development pipeline?

Your question might be covered by the following feature request: Support FC HBA's on *nix Platforms (here)

A scheduling mechanism with the flexibility of "cron" would be extremely useful. Eg:Schedules such as: * Every hour, between 9am to 5pm, on week days. * 30 minutes past every 6th hour (eg: 0:30, 6:30, 12:30, 18:30). * Every 10 minutes, to update a "Publish Report (Overwrite)" image, for display elsewhere via image URL. -…