Azure Java SDK reaction to connection resets when using Azure blob

[mahmoudsamy27]

application based on Azure Java SDK is getting intermittent expected connection resets while communicating with azure blob storage.

the application crashes and not retrying the error properly. Here is the error:

--- [or-http-epoll-1] r.netty.http.client.HttpClientConnect : [5303716f-219, L:/10.x.x.x:59338 - R:removedthename.blob.core.windows.net/] The connection observed an error, the request cannot be retried as the headers/body were sent

->: io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer

Ask:

• Does Azure Java SDK for storage retires “connection reset by peer” errors as above?
• Are there any recommendations to handle these exceptions gracefully. Found a lead here but it is for netty Connection reset by peer exception reactor/reactor-netty#1774

[alzimmermsft]

Thanks for filing this @mahmoudsamy27.

There is this PR (#32648) which is improving inspection of errors to determine what should be retried. Though in the case you reported, is the NativeIoException being thrown or is it just being logged? If it's being thrown the PR I mentioned should help alleviate the request not being retried.

For the log message, this is something the Azure SDKs can't control as it's logged by Reactor Netty/Netty in their code and logging is determined by customer logging configurations.

[ghost]

Hi, we're sending this friendly reminder because we haven't heard back from you in a while. We need more information about this issue to help address it. Please be sure to give us your input within the next 7 days. If we don't hear back from you within 14 days of this comment the issue will be automatically closed. Thank you!