-
-
Notifications
You must be signed in to change notification settings - Fork 3.7k
[linktap] Add communication timeout configuration parameter #18679
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
[linktap] Addition of timeout config option Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Cleanup of unrequired variable Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Revert accidental removal of config code Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Remove commented out line Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Align const to final setting name Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Cleanup to use handler function Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Comment for magic number Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Revert i18n file Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Missed call remap to single method Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Remove unnecessary parameter addition Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] Spotless apply Signed-off-by: David Goodyear <david.goodyear@gmail.com>
[linktap] i18n additions for new config Signed-off-by: David Goodyear <david.goodyear@gmail.com>
This pull request has been mentioned on openHAB Community. There might be relevant details there: https://community.openhab.org/t/binding-request-linktap/77192/48 |
bundles/org.openhab.binding.linktap/src/main/resources/OH-INF/thing/thing-types.xml
Outdated
Show resolved
Hide resolved
Technically the changes are in my branch and tested - seems to be having issues syncing the PR request with the final commits. Will check tomorrow the state of play - to see if its up to date, so I can remove the draft status and WIP from the title. Apologies I only just saw there was a work in progress label now :( |
[linktap] force push Signed-off-by: David Goodyear <david.goodyear@gmail.com>
88e1239
to
51b3cf5
Compare
[linktap] adjust scope to private Signed-off-by: David Goodyear <david.goodyear@gmail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks. Two minor comments left, otherwise LGTM.
bundles/org.openhab.binding.linktap/src/main/resources/OH-INF/i18n/linktap.properties
Outdated
Show resolved
Hide resolved
[linktap] missed renames Signed-off-by: David Goodyear <david.goodyear@gmail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@jlaur Thank you for the quick turnaround 💯 👍 |
…18679) Signed-off-by: David Goodyear <david.goodyear@gmail.com>
This is to allow users to adjust the timeout when responses are required to compensate either for slow gateway's or starvation on the runtime. This relates to the last few discussion points around 11/05 here: https://community.openhab.org/t/binding-request-linktap/77192/44.
From tracing the software through, it can only be I suspect either starvation of jetty being able to service the request's in the time given when the system is under full load likely based on how the CPU's are thread balancing on PI hardware, or the older generation of device isn't responding. (I doubt this is the case as generally the GW's have been very stable).
This therefore adds a advanced option, where users can increase the timeout to be between 3 - 30 seconds for how long the binding will wait for the underlying jetty to have capacity to process the response.
If trace mode is enabled jetty requests are traced - to check for starvation (e.g. never exit the queued state). This is so its clear if other bindings are locking the http client are potentially starving the binding from being able to execute http requests. The impact of the jetty traces should be minimal overall. It will only go through those flows during the initial connection setup, or when commands are issued. In normal operation the GW send's data to the servlet which isn't logging more information, only if it was operating in the failsafe polling mode would it potentially be adding a bit more trace data.