KEYCLOAK-4278 Testsuite failures with -Pauth-server-wildfly

[mstruk]

• Fixed timeouts in tests involving sending of emails
• Excluded one test requiring auth-server-undertow from executing as part of auth-server-wildfly

[stianst]

There is a separate JIRA to fix this test. We should not exclude it like this, but rather the test needs fixing.

[mstruk]

@stianst True, that part is a temporary solution. I can remove it.

[stianst]

Please remove it

[mstruk]

It's been removed.

[stianst]

Actually, I don't think this is the correct approach at all. Graphene has different waits depending of the type of request. The problem isn't the "page load" timeout. It's that the tests are using the wrong timeout.

@pdrozd can you confirm?

[pdrozd]

@mstruk The default page load timeout is set to 10s and I think this is sufficient for almost all the tests. Are these tests fail on your local machine? On jenkins the tests are passing.
If these tests are failing regularly, there should be different problem, which needs to be solved rather than increasing the page load timeout.

[mstruk]

On my machine these tests actually take longer than 10 secs to complete when auth-server-wildfly is activated. I couldn't quite get to the bottom of it but I suspect some lazy init of some provider to do with sending secured emails might be to blame. I can try and explore some more to better understand where the time is getting lost.

[mstruk]

More specifically, waiting for validate email required action which renders and email, and sends it as part of processing of HTTP request takes longer than 10 seconds. Not sure why.

[stianst]

Something truly weird must be happening on your machine @mstruk. I can't see how it should take 10+ seconds to send en email with a test server.

[mstruk]

Whatever it is, it only causes tests that send emails to fail.
I'll do some profiling to pin down the root cause.

[mstruk]

The problem is apparently incredibly long local hostname resolution on my system:

   java.lang.Thread.State: RUNNABLE
	at java.net.Inet4AddressImpl.getLocalHostName(Native Method)
	at java.net.InetAddress.getLocalHost(InetAddress.java:1474)
	at javax.mail.internet.InternetAddress.getLocalHostName(InternetAddress.java:568)
	at javax.mail.internet.InternetAddress._getLocalAddress(InternetAddress.java:547)
	at javax.mail.internet.InternetAddress.getLocalAddress(InternetAddress.java:518)
	at javax.mail.internet.UniqueValue.getUniqueMessageIDValue(UniqueValue.java:100)
	at javax.mail.internet.MimeMessage.updateMessageID(MimeMessage.java:2172)
	at javax.mail.internet.MimeMessage.updateHeaders(MimeMessage.java:2200)
	- locked <0x00000007bc1ac900> (a com.sun.mail.smtp.SMTPMessage)
	at javax.mail.internet.MimeMessage.saveChanges(MimeMessage.java:2159)
	at org.keycloak.email.DefaultEmailSenderProvider.send(DefaultEmailSenderProvider.java:132)
	at org.keycloak.email.freemarker.FreeMarkerEmailTemplateProvider.send(FreeMarkerEmailTemplateProvider.java:191)

   java.lang.Thread.State: RUNNABLE
	at java.net.Inet4AddressImpl.getLocalHostName(Native Method)
	at java.net.InetAddress.getLocalHost(InetAddress.java:1474)
	at com.sun.mail.smtp.SMTPTransport.getLocalHost(SMTPTransport.java:255)
	- locked <0x00000007bc3756f8> (a com.sun.mail.smtp.SMTPTransport)
	at com.sun.mail.smtp.SMTPTransport.protocolConnect(SMTPTransport.java:703)
	- locked <0x00000007bc3756f8> (a com.sun.mail.smtp.SMTPTransport)
	at javax.mail.Service.connect(Service.java:366)
	- locked <0x00000007bc3756f8> (a com.sun.mail.smtp.SMTPTransport)
	at javax.mail.Service.connect(Service.java:246)
	at javax.mail.Service.connect(Service.java:195)
	at org.keycloak.email.DefaultEmailSenderProvider.send(DefaultEmailSenderProvider.java:139)
	at org.keycloak.email.freemarker.FreeMarkerEmailTemplateProvider.send(FreeMarkerEmailTemplateProvider.java:191)

I found more info and a solution.

I'm closing this issue.