cacti icon indicating copy to clipboard operation
cacti copied to clipboard

Button to reindex bad indexes

Open bmfmancini opened this issue 6 years ago • 3 comments

Hey Guys

One feature that would be awsome to have is a way to re-index devices with reported bad indexes

we have outputs in the log that show something like this

WARNING: You have 145 Devices with bad SNMP Indexes. Devices: Device[1210], Device[1215]

But it would be so useful if these 145 devices could be re-indexed to help solve the problem the only other way I know to re-index them would be either one by one or all of the devices I am sure you could also do some cli magic with the poller_reindex script as well but it would be cool to be able to do it via the gui

bmfmancini avatar Dec 04 '19 18:12 bmfmancini

Yea, this should go on the System Utilities page. You are using the 1.2.x branch right? There is a force re-index for all hosts, but that will rerun for all devices. You should really troubleshoot why those bad indexes happened. Could be a crap data query (like Linux processes), or you simply have ifNames that are going away. The seconds not likely unless you are removing or replacing cards with differing port counts in your switches.

cigamit avatar Dec 06 '19 01:12 cigamit

yea the devices with the bad indexs are a small portion of the total devices they are older ciscos some on older IOS as well so I am sure its all sorts of issues

bmfmancini avatar Dec 06 '19 01:12 bmfmancini

With the new re-indexing in 1.2, this could be pretty fast. But the screen would likely hang for a while unless it's done in background, which means you need a status page too, like boost.

TheWitness avatar Mar 21 '21 21:03 TheWitness

This was actually resolved some time ago. I'll mark resolved for 1.2.23 though.

TheWitness avatar Aug 18 '22 11:08 TheWitness