forked from docs/doc-exports
Chen, Junjie
ca9677b121
Reviewed-by: Antonova, Ekaterina <ekantono@noreply.gitea.eco.tsi-dev.otc-service.com> Co-authored-by: Chen, Junjie <chenjunjie@huawei.com> Co-committed-by: Chen, Junjie <chenjunjie@huawei.com>
13 lines
1.0 KiB
HTML
13 lines
1.0 KiB
HTML
<a name="kafka_faq_0056"></a><a name="kafka_faq_0056"></a>
|
|
|
|
<h1 class="topictitle1">Why Does Message Production Fail During Scaling?</h1>
|
|
<div id="body0000001562223800"><p id="kafka_faq_0056__p8060118"><strong id="kafka_faq_0056__b134691237152619">Possible cause</strong>: When you increase the broker flavor, a rolling restart is performed on brokers. During the restart, partition leaders are changed. The producer has cached the old partition leader IDs, so messages are still sent to old partition leaders. As a result, messages fail to be produced.</p>
|
|
<p id="kafka_faq_0056__p1626514253489"><strong id="kafka_faq_0056__b1192034122611">Solution</strong>: Configure the retry mechanism on the producer by setting <strong id="kafka_faq_0056__b1967719911273">retries</strong> to <strong id="kafka_faq_0056__b99261212152713">Integer.MAX_VALUE</strong>.</p>
|
|
</div>
|
|
<div>
|
|
<div class="familylinks">
|
|
<div class="parentlink"><strong>Parent topic:</strong> <a href="kafka_faq_0053.html">Specification Modification</a></div>
|
|
</div>
|
|
</div>
|
|
|