Dmitriy, Check up Ray’s initial message. There is no single cache.get command or SQL query that can trigger data movement from disk to RAM. Plus, the output generated after the restart suggests that the node found all the persistence directories. Thus, not sure how this issue is related to consistent ID mess discovered by Yakov before.
Ray, please chime in and clarify if you tried to work with the data after the restart. — Denis > On Sep 15, 2017, at 4:02 PM, Dmitry Pavlov <dpavlov....@gmail.com> wrote: > > Hi Denis, > > I agree that data presence should be double-checked. In case check will show > that 1) cache doesn't exist at all after restart and 2) we can find two > folders with binary_meta and wal that means this question is similar with > recent question > http://apache-ignite-users.70518.x6.nabble.com/Specifying-location-of-persistent-storage-location-tc16636.html > > <http://apache-ignite-users.70518.x6.nabble.com/Specifying-location-of-persistent-storage-location-tc16636.html> > > > It is completely transparent to user that data is in RAM or on disk. > > > Hi Yakov, > > To my mind ticket is reasonable. I can suggest to generate such warning only > in case consistent ID is not overriden by IgniteConfiguration. Parameter > override may indicate that several nodes running at the same machine. > > Sincerely, > Dmitriy Pavlov > > сб, 16 сент. 2017 г. в 1:47, Denis Magda <dma...@apache.org > <mailto:dma...@apache.org>>: > Alexey, > > The data preloading from disk to RAM doesn’t occur automatically after the > restart. It will be moved to RAM on demand while your application executes > its logic. > > Ray, > > Have you executed any query (key-value, SQL) to confirm that the data is > missing after the shutdown? > > — > Denis > > > On Sep 15, 2017, at 4:09 AM, Alexey Kukushkin <kukushkinale...@gmail.com > > <mailto:kukushkinale...@gmail.com>> wrote: > > > > Ray, > > > > Yes, enabling persistence must automatically bring all your data and > > metadata upon the cluster startup. Please give us more details why you > > think the cache was not created. BTW, do you activate the cluster after > > startup? Enabling persistence requires explicit cluster activation: > > > > $IGNITE_HOME/bin/control.sh --host 127.0.0.1 --port 11211 --activate > > > > >
