Fatal Exception: java.lang.OutOfMemoryError

[ranjeet-porter]

Fatal Exception: java.lang.OutOfMemoryError: pthread_create (1040KB stack) failed: Try again
at java.lang.Thread.nativeCreate(Thread.java)
at java.lang.Thread.start(Thread.java:733)
at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.scheduleReconnect(AWSIotMqttManager.java:1377)
at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.handleConnectionFailure(AWSIotMqttManager.java:1356)
at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.reconnectWithIAM(AWSIotMqttManager.java:1346)
at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.reconnectToSession(AWSIotMqttManager.java:1278)
at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager$4.run(AWSIotMqttManager.java:1385)
at android.os.Handler.handleCallback(Handler.java:873)
at android.os.Handler.dispatchMessage(Handler.java:99)
at android.os.Looper.loop(Looper.java:201)
at android.os.HandlerThread.run(HandlerThread.java:65)

SDK Version : com.amazonaws:aws-android-sdk-iot: 2.16.12
Platform : Android

[lawmicha]

Hi @ranjeet-porter, can you provide us with some more details around how to reproduce this with some code snippets, call pattern, and set up? Out of memory exception could be caused by different processes, were you able to profile your app during the crash to identify what's using up the memory?

[ranjeet-porter]

Hi @lawmicha
Sure. I have setup my mqtt re-connection with -1 mean infinite times so it is calling reconnection multiple times within second then causing out of memory after certain times.

Fatal Exception: java.lang.OutOfMemoryError: pthread_create (1040KB stack) failed: Try again at java.lang.Thread.nativeCreate(Thread.java) at java.lang.Thread.start(Thread.java:733) at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.scheduleReconnect(AWSIotMqttManager.java:1377) at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.handleConnectionFailure(AWSIotMqttManager.java:1356) at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.reconnectWithIAM(AWSIotMqttManager.java:1346) at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager.reconnectToSession(AWSIotMqttManager.java:1278) at com.amazonaws.mobileconnectors.iot.AWSIotMqttManager$4.run(AWSIotMqttManager.java:1385) at android.os.Handler.handleCallback(Handler.java:873) at android.os.Handler.dispatchMessage(Handler.java:99) at android.os.Looper.loop(Looper.java:201) at android.os.HandlerThread.run(HandlerThread.java:65)

Device Specification:

Device
Brand:Xiaomi
Model:Redmi 8A Dual
Orientation:Portrait
RAM free: 933.63 MB
Disk free: 5.14 GB
Operating System
Version:Android 9
Orientation:Portrait
Rooted: No
Crash
Date:Jul 20, 2024, 8:31:33 PM
App version:5.101.4 (371)
User
ID:9767175862620

[lawmicha]

could you share the code snippets for this?

I have setup my mqtt re-connection with -1

[ranjeet-porter]

AWSIotMqttManager(mqttConfig.clientId, mqttConfig.awsIotEndPoint).apply {
    mqttConfig.mqttClientConfig?.let {
        keepAlive = it.keepAlive
        isAutoReconnect = it.isAutoReconnect
        isOfflinePublishQueueEnabled = it.isOfflinePublishQueueEnabled
        maxAutoReconnectAttempts = -1
        setReconnectRetryLimits(
            it.minReconnectRetryTime,
            it.maxReconnectRetryTime,
        )
    }
}

[tylerjroach]

@ranjeet-porter

I'm a bit confused by

Sure. I have setup my mqtt re-connection with -1 mean infinite times so it is calling reconnection multiple times within second then causing out of memory after certain times.

What values have you set for setReconnectRetryLimits minTimeout and maxTimeout (Default is 4 seconds and 64 seconds). Those values are set in seconds so I wouldn't expect a retry to be attempted at more than once per second. The retries should then exponentially backoff to the maxTimeout value set there.

Are you setting both of those values to 0?

[ranjeet-porter]

I set minTimeout = 4 and maxTimeout = 15 but with maxAttempt is -1 and I got out of memory exception

[tylerjroach]

@ranjeet-porter It would be helpful if you were able to add logs of a connect attempt session. I believe you can enable in the SDK by simply calling LogFactory.setLevel(LogFactory.Level.DEBUG);.

Also curious if you have tried a high (but not infinite value). Ex: maxAttempt = 100. I'm curious if you still see the same behavior.

In a quick glance at the code, the exponential backoff should prevent too many attempts, but to receive OOM exceptions seems to indicate that reconnects are rapidly being attempted.