Channels
doorman
zercurity
infrastructure
code-review
queryhub
apple-silicon
carving
goquery
aws
querycon
golang
file-carving
fuzzing
help-proxy
darkbytes
process-auditing
general
windows
random
fleet-dev
tls
fim
zentral
zeek
auditing-warroom
linen-dev
fleetosquery
plugins
jobs
arm-architecture
uptycs
android_tests
selfgroup
vendor-feeds
fleet
eclecticiq-polylogyx-extension
ebpf
website
core
macos
kolide
osctrl
extensions
foundation
sql
officehours
linux
community-feeds
Powered by
#fleet
Title
z
zwass
12/23/2020, 9:56 PM
Are you seeing that same error message? Do you see the same value for
host_identifier
in the log line?
More details in
https://github.com/fleetdm/fleet/issues/102
. If you are seeing a perf issue please follow these instructions to get a debug archive that we can use to debug the perf:
https://github.com/fleetdm/fleet/blob/master/docs/infrastructure/performance.md#generate-debug-archive-fleet-340
c
Chris Reisor
12/23/2020, 10:02 PM
ah, we're on 3.1.0 😞
z
zwass
12/23/2020, 10:03 PM
We've only seen this in association with the duplicate host identifier problem, so consider whether you could be running into that scenario. If you are able to modify the osqueryd flagfile, setting
host_identifier=instance
can be a good remediation.
c
Chris Reisor
12/23/2020, 10:05 PM
OK thanks, I'll look into that
Do you know if MySQL should have
innodb_rollback_on_timeout
enabled?
z
zwass
12/24/2020, 12:38 AM
Fleet's tx handling (
https://github.com/fleetdm/fleet/blob/master/server/datastore/mysql/mysql.go#L63-L99
) will rollback on any error encountered during the transaction, so I don't think that should make any difference. Please let us know if you observe any issues.
2 Views
Post