Christian- Yeah, you are hitting some of the same issues I did. There doesn't seem to be a straight-line path for supporting TCP connection health in a distributed environment. I started making some patches to improve H2's TCP client management.. there is still open discussion on the approach and need (https://github.com/h2database/h2database/pull/1990).
Q: What is the use case where you set AUTOCOMMIT=FALSE on the url? It seems to me you would want AUTO_RECONNECT. Specifically-- would your use case instead support disabling autoCommit on the connection object and deal with the in doubt transactions instead? -Matt On Monday, September 9, 2019 at 9:39:05 AM UTC-5, Christian Buchegger wrote: > > Matt- > thanks for your suggestions. Unfortunate as our application requires > AUTOCOMMIT=FALSE; AUTO_RECONNECT is not an option. > Also this would just help for situations when a reconnect succeeds, and > still leave stale sessions. > > I am aware of cancel_session(). It is documented as "Cancels the currently > executing statement of another session." and exactly behaves as such. > The session will still exist after using this function. > > Thanks, > Christian > > Am Montag, 9. September 2019 15:14:19 UTC+2 schrieb Matt Pavlovich: >> >> Look into setting these combinations: >> >> 1. On the server: h2.socketConnectTimeout >> 2. On the client urls: ;AUTO_RECONNECT=TRUE >> >> The behavior will be that clients re-connect over and over (harmless, but >> not ideal) but you won’t have stale connections or abandoned locks. >> >> > >> Also note— there already exists a function to close sessions: >> >> http://www.h2database.com/html/functions.html#cancel_session >> >> -Matt >> >> -- You received this message because you are subscribed to the Google Groups "H2 Database" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/h2-database/4567a274-fb30-4f22-8491-8b3286ea43c0%40googlegroups.com.
