Hi, GridCacheTtlManager.expire is called from different places and may lead to cluster hanging. It is impossible to understand from user code was the operation successful or not when such NPE was thrown.
As this method is called (from gateway.leave() method) after operation has been finished, so, your transaction has been commited successfully. On Fri, Apr 13, 2018 at 8:31 AM, Bellenger, Dominique < dominique.bellen...@ipdynamics.de> wrote: > Thank you for the information! What are the worst consequences if I hit > that bug? Entries not being evicted? > Is my tx committed if the error is striking on commit or do I need to redo > the work that was done in the tx? > > > On 12. Apr 2018, at 14:30, Andrey Mashenkov <andrey.mashen...@gmail.com> > wrote: > > Hi Dome, > > It is known issue and there is a ticket for this [1] you can track. > > It is hard to be reproduced due to a race. > Seems, GridCacheUtils.unwindEvicts() should check if context has been > started. > > > [1] https://issues.apache.org/jira/browse/IGNITE-7972 > > > On Thu, Apr 12, 2018 at 2:39 PM, dkarachentsev <dkarachent...@gridgain.com > > wrote: > >> Hi Dome, >> >> Could you please attach full logs? >> >> Thanks! >> -Dmitry >> >> >> >> -- >> Sent from: http://apache-ignite-users.70518.x6.nabble.com/ >> > > > > -- > Best regards, > Andrey V. Mashenkov > > -- Best regards, Andrey V. Mashenkov
