Re: [keycloak-dev] Potential performance regression between 3.4.3.Final and 4.3.0.Final

Thanks Hynek, I’ll test this. I also tracked the issue down to the ModelToRepresentation.toRepresentation(KeycloakSession session, RealmModel realm, UserModel user) method. However, what I am also seeing is when the method is running, it is emitting SQL queries that are not scoped down to the user id of the user being transformed. E.g. for the user attributes of a user the following query is generated: select attributes0_.USER_ID as USER_ID4_72_1_, attributes0_.ID as ID1_72_1_, attributes0_.ID as ID1_72_0_, attributes0_.NAME as NAME2_72_0_, attributes0_.USER_ID as USER_ID4_72_0_, attributes0_.VALUE as VALUE3_72_0_ from USER_ATTRIBUTE attributes0_ where attributes0_.USER_ID in ( select userentity0_.ID from USER_ENTITY userentity0_ where userentity0_.REALM_ID=? and ( userentity0_.SERVICE_ACCOUNT_CLIENT_LINK is null ) ) The subselect returns all real users from the realm. It is neither paged nor bound to a specific user id. The same happens in the above method for credentials and required actions. I am not a JPA/Hibernate expert so I might be missing something here… Best regards, Sebastian Mit freundlichen Grüßen / Best regards *Dr.-Ing. Sebastian Schuster * Engineering and Support (INST/ESY1) Bosch Software Innovations GmbH | Ullsteinstr. 128 | 12109 Berlin | GERMANY | www.bosch-si.com Tel. +49 30 726112-485 | Fax +49 30 726112-100 | Sebastian.Schuster(a)bosch-si.com Sitz: Berlin, Registergericht: Amtsgericht Charlottenburg; HRB 148411 B Aufsichtsratsvorsitzender: Dr.-Ing. Thorsten Lücke; Geschäftsführung: Dr. Stefan Ferber, Michael Hahn *From:* Hynek Mlnarik <hmlnarik(a)redhat.com&gt; *Sent:* Mittwoch, 29. August 2018 16:35 *To:* Thomas Darimont <thomas.darimont(a)googlemail.com&gt; *Cc:* Stian Thorgersen <stian(a)redhat.com>; keycloak-dev < keycloak-dev(a)lists.jboss.org>; Schuster Sebastian (INST/ESY1) < Sebastian.Schuster(a)bosch-si.com&gt; *Subject:* Re: [keycloak-dev] Potential performance regression between 3.4.3.Final and 4.3.0.Final Thanks Thomas and all, great help with reproducing the issue at least for admin user list endpoint. Could you please test [1] that it indeed fixes the issue? It was caused obtaining too much user details when unnecessary (for user list, only few fields are necessary). As for increased memory requirements, these might be unrelated to this fix, and we'll keep an eye on this. --Hynek [1] https://github.com/keycloak/keycloak/pull/5517 On Mon, Aug 27, 2018 at 4:48 PM Thomas Darimont < thomas.darimont(a)googlemail.com&gt; wrote: Hello, FYI I created https://issues.jboss.org/browse/KEYCLOAK-8150 Cheers, Thomas Am Mo., 27. Aug. 2018 um 13:33 Uhr schrieb Stian Thorgersen < sthorger(a)redhat.com&gt;: > Can you open a bug for this in JIRA please and include as much details as > you can. > > On Thu, 23 Aug 2018 at 15:45, Thomas Darimont < > thomas.darimont(a)googlemail.com&gt; wrote: > >> Hi Sebastian, >> >> thanks for investigating! Unfortunately, I didn't have time yet to look >> further into the problem, but I also observed the slow attributes handling. >> Btw. we have around 100k users and ~700k attributes. The >> Entity-Attribute-Value data model takes it's toll here... >> >> @Stian so far I've only observed this when scrolling through the user >> overview in the admin-console, which also triggers >> the mentioned spike in heap usage. >> >> Cheers, >> Thomas >> >> Am Do., 23. Aug. 2018 um 15:22 Uhr schrieb Schuster Sebastian (INST/ESY1) >> <Sebastian.Schuster(a)bosch-si.com&gt;: >> >>> Btw. we also have 40K user attributes... >>> >>> Best regards, >>> Sebastian >>> >>> Mit freundlichen Grüßen / Best regards >>> >>> Dr.-Ing. Sebastian Schuster >>> >>> Engineering and Support (INST/ESY1) >>> Bosch Software Innovations GmbH | Ullsteinstr. 128 | 12109 Berlin | >>> GERMANY | www.bosch-si.com >>> Tel. +49 30 726112-485 | Fax +49 30 726112-100 | >>> Sebastian.Schuster(a)bosch-si.com >>> >>> Sitz: Berlin, Registergericht: Amtsgericht Charlottenburg; HRB 148411 B >>> Aufsichtsratsvorsitzender: Dr.-Ing. Thorsten Lücke; Geschäftsführung: >>> Dr. Stefan Ferber, Michael Hahn >>> >>> >>> >>> >>> -----Original Message----- >>> From: keycloak-dev-bounces(a)lists.jboss.org < >>> keycloak-dev-bounces(a)lists.jboss.org&gt; On Behalf Of Stian Thorgersen >>> Sent: Mittwoch, 22. August 2018 21:25 >>> To: Thomas Darimont <thomas.darimont(a)googlemail.com&gt; >>> Cc: keycloak-dev <keycloak-dev(a)lists.jboss.org&gt; >>> Subject: Re: [keycloak-dev] Potential performance regression between >>> 3.4.3.Final and 4.3.0.Final >>> >>> That's quite worrying. Is it limited to browsing users through the admin >>> console or are you seeing bad behaviour elsewhere? Same question applies to >>> the heap. My best bet here is that it has that something has changed around >>> user querying/caching. >>> >>> On Mon, 20 Aug 2018 at 19:18, Thomas Darimont < >>> thomas.darimont(a)googlemail.com&gt; wrote: >>> >>> > Hello Keycloak team, >>> > >>> > has anyone encountered some performance issues after upgrading 3.4.3 >>> > to 4.x (4.3.0)? >>> > >>> > Today I noticed a performance regression while preparing an upgrade >>> > from Keycloak 3.4.3.Final to 4.3.0.Final in our staging environment. >>> > >>> > In our test environment, we have around ~100k test users stored in a >>> > postgres-backed database. When we started the server with the new >>> > Keycloak version, the migration went through, and everything looked >>> > fine at first, but when we tried to browse the list of users via the >>> > admin-console, we noticed that the CPU and memory consumption of the >>> > server increased significantly, up to a point where Keycloak crashed >>> with an OOME. >>> > >>> > All previous Keycloak versions including 3.4.3 were very modest with >>> > their memory requirements and quite happy with ~1g heap. >>> > However, that seems to have changed in Keycloak 4.3.0 - there we >>> > needed at least 4g to prevent Keycloak from crashing with an OOME. >>> > Furthermore, we noticed that the response times for browsing the >>> > paginated user view increased significantly as well: >>> > In Keycloak 3.4.3 the average time to load a user page is ~80ms. In >>> > Keycloak 4.3.0 (and older versions >= 4.0.0.Beta1) the same operation >>> > takes >>> > ~7 seconds for a test realm with just 10k users. >>> > >>> > In the test realm with 100k users, the time to load a single page in >>> > the users listing was 66 seconds for version 4.3.0, on average - >>> > compared to quite stable 80ms in 3.4.3. >>> > >>> > The database query that is executed by Keycloak 4.3.0 runs in ~1.5 >>> > seconds for 100k users, so I assume the processing logic in Keycloak >>> > is the culprit. >>> > >>> > The problem of long load-times can be reproduced with the Keycloak >>> > docker images and the in-memory database. I also created a small >>> > example project that creates some users with just a few attributes in >>> > a docker based 3.4.3 and 4.3.0 Keycloak environment with a Postgres >>> > database to reproduce the problem. >>> > https://github.com/thomasdarimont/kc-user-regression-tester >>> > >>> > Cheers, >>> > Thomas >>> > _______________________________________________ >>> > keycloak-dev mailing list >>> > keycloak-dev(a)lists.jboss.org >>> > https://lists.jboss.org/mailman/listinfo/keycloak-dev >>> > >>> _______________________________________________ >>> keycloak-dev mailing list >>> keycloak-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/keycloak-dev >>> >> _______________________________________________ keycloak-dev mailing list keycloak-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/keycloak-dev