This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Created attachment 122847 [details] the thread dump after opening a JPA entity in the editor We have a large J2EE project with hundreds of JPA entities. When I open the source for an entity (with 8 named queries in it) Netbeans process goes from 700Mb to 4Gb of allocated memory and then stays up indefinitely. Notice that I don't do anything in the editor, just keeping that only file open. To achieve this I increased the heap space to 4Gb; keeping the default settings prevents me to open the file at all (Netbeans freezes and I need to kill the process). you can find the thread dump attached thanks
looks like a duplicate of issue 216053 but provide some hint why actions may be slow, may be because of memory consumption.
snapshot show about 2000 calls to suspected in slowness method, it may be the reason for temporary get a lot of memory as if I got it right there is no oom even with initial heap size, just gc can't be fast enouth. but also it show it's duplicate of issue 216053.can you try latest dev build and create snapshot once again?can you provide jpql query sample from you entity is you can share any? *** This bug has been marked as a duplicate of bug 216053 ***
here is a sample of the queries: @NamedQuery(name = "ValoreIndicatore.findMostRecentInContesto", query = "SELECT v FROM ValoreIndicatore AS v LEFT JOIN v.versionePerGruppo.tipo LEFT JOIN v.istanzaIndicatore.pagina.scheda.modelloValutazione.nodi as nodo " + "WHERE v.cartella.id=:cartellaId AND " + "nodo.id=:nodoId AND v.istanzaIndicatore.pagina.scheda.modelloValutazione.tipoSezione.id=3 " + "AND (v.versionePerGruppo.gruppo.id=:groupId OR v.versionePerGruppo.tipo.tipoVersioneSistema.id != 1) AND v.versionePerGruppo.data = " + "(SELECT MAX(v1.versionePerGruppo.data) FROM ValoreIndicatore as v1 LEFT JOIN v1.versionePerGruppo.tipo left join v1.istanzaIndicatore.pagina.scheda.modelloValutazione.nodi as nodo " + "WHERE v1.cartella.id=:cartellaId " + "AND nodo.id=:nodoId AND v1.istanzaIndicatore.pagina.scheda.modelloValutazione.tipoSezione.id=3 " + "AND (v1.versionePerGruppo.gruppo.id=:groupId OR v1.versionePerGruppo.tipo.tipoVersioneSistema.id != 1))"), @NamedQuery(name = "ValoreIndicatore.findByModello", query = "SELECT v FROM ValoreIndicatore AS v WHERE v.istanzaIndicatore.pagina.scheda.modelloValutazione.id = :modValId " + "AND v.versionePerGruppo.id IN (:versionIds)"), @NamedQuery(name = "ValoreIndicatore.findByVersion", query = "SELECT v FROM ValoreIndicatore AS v WHERE v.versionePerGruppo.id = :versionId") I am downloading the nightly build, will post the new thread dump asap
can you provide a hint how this v1.istanzaIndicatore.pagina.scheda.modelloValutazione.tipoSezione.id may be constructed, if v1 is some entity, what is istanzaIndicatore.pagina.scheda.modelloValutazione.tipoSezione.id? it may help to determine this case and skip some long action for this.
it's a chain of @ManyToOne relationships between entities ValoreIndicatoreBase.java: ... @ManyToOne @JoinColumn(name = "ID_TBL_ISTANZA_INDICATORE", referencedColumnName = "ID", nullable = true) public IstanzaIndicatore getIstanzaIndicatore() { return istanzaIndicatore; } ... IstanzaIndicatore.java: ... @ManyToOne @JoinColumn(name = "ID_pagina", referencedColumnName = "IDPAGINA", nullable = true) public Pagina getPagina() { return pagina; } ... and so on
Created attachment 122881 [details] thread dump with 201208070001 build
I tried the 201208070001 build (find the dump attached). It behaves better, byt memory still goes up to 1.9Gb by only opening the entity source code (memory then stays steady at 1.9). Editor becomes less responsive, and autocomplete for named queries needs a very long time to show its suggestions.
thanks, I see calls to Elements.getTypeElement drop from 2000 to 250 in this build, is it with the same entity? Also as I got it's not enough so far.
yes it's the same entity as before