On 04/22/2017 11:54 PM, Bill Sirinek wrote:
A side effect of this change is that my PuppetDB command queue depth
went from just over 100 to 0, and has stayed there for a couple hours
now, maybe occasionally poking up to 2-3 briefly. At least I think it
was related to this change. :)
You're pr
Thanks Wyatt, that worked! The alter table command returned immediately. I
only have about 170k reports in my DB since I migrated to 2017.1.1 friday
night. I run puppet hourly on about 8000 nodes.
A side effect of this change is that my PuppetDB command queue depth went
from just over 100 to 0,
On 04/22/2017 10:35 AM, Bill Sirinek wrote:
2017-04-22 13:30:25.912 EDT
[db:pe-puppetdb,sess:58fb8caf.8cbb,pid:36027,vtid:29/23277,tid:396758]
STATEMENT: INSERT INTO resource_events ( new_value,
corrective_change, property, file, report_id, old_value,
containing_class, certname_id, line,
I have a module that manages various bios settings on HP proliant servers
(basically hacked up the conrep module from the forge) Anyhow, the resource
names (resource_title) can get fairly long and break the 40 character
column width in the database, as shown with the error below:
2017-04-22 13
