Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[platform-dev] Concerns about org.eclipse.ecf.provider.filetransfer.httpclient45

Hi,

I don't know how many of you have noticed things like this in the build logs:

12:25:45 Mar 04, 2020 11:25:45 AM org.apache.http.impl.execchain.RetryExec execute
12:25:45 INFO: I/O exception (org.apache.http.NoHttpResponseException) caught when processing request to {s}->https://download.eclipse.org:443: The target server failed to respond
12:25:45 Mar 04, 2020 11:25:45 AM org.apache.http.impl.execchain.RetryExec execute
12:25:45 INFO: Retrying request to {s}->https://download.eclipse.org:443

I'm a bit concerned about the cause of such exceptions.  I've debugged where this happens in my Oomph environment.  Specifically it happens in org.apache.http.impl.execchain.RetryExec.execute(HttpRoute, HttpRequestWrapper, HttpClientContext, HttpExecutionAware) when parsing the response header finds nothing with the following stack leading to the logged problem:

org.apache.http.NoHttpResponseException.<init>(java.lang.String) line: 47   
org.apache.http.impl.conn.DefaultHttpResponseParser.parseHead(org.apache.http.io.SessionInputBuffer) line: 141   
org.apache.http.impl.conn.DefaultHttpResponseParser.parseHead(org.apache.http.io.SessionInputBuffer) line: 56   
org.apache.http.impl.conn.DefaultHttpResponseParser(org.apache.http.impl.io.AbstractMessageParser<T>).parse() line: 259   
org.apache.http.impl.conn.LoggingManagedHttpClientConnection(org.apache.http.impl.DefaultBHttpClientConnection).receiveResponseHeader() line: 163   
org.apache.http.impl.conn.CPoolProxy.receiveResponseHeader() line: 157   
org.apache.http.protocol.HttpRequestExecutor.doReceiveResponse(org.apache.http.HttpRequest, org.apache.http.HttpClientConnection, org.apache.http.protocol.HttpContext) line: 273   
org.apache.http.protocol.HttpRequestExecutor.execute(org.apache.http.HttpRequest, org.apache.http.HttpClientConnection, org.apache.http.protocol.HttpContext) line: 125   
org.apache.http.impl.execchain.MainClientExec.execute(org.apache.http.conn.routing.HttpRoute, org.apache.http.client.methods.HttpRequestWrapper, org.apache.http.client.protocol.HttpClientContext, org.apache.http.client.methods.HttpExecutionAware) line: 272   
org.apache.http.impl.execchain.ProtocolExec.execute(org.apache.http.conn.routing.HttpRoute, org.apache.http.client.methods.HttpRequestWrapper, org.apache.http.client.protocol.HttpClientContext, org.apache.http.client.methods.HttpExecutionAware) line: 186   
org.apache.http.impl.execchain.RetryExec.execute(org.apache.http.conn.routing.HttpRoute, org.apache.http.client.methods.HttpRequestWrapper, org.apache.http.client.protocol.HttpClientContext, org.apache.http.client.methods.HttpExecutionAware) line: 89   
org.apache.http.impl.execchain.RedirectExec.execute(org.apache.http.conn.routing.HttpRoute, org.apache.http.client.methods.HttpRequestWrapper, org.apache.http.client.protocol.HttpClientContext, org.apache.http.client.methods.HttpExecutionAware) line: 110   
org.apache.http.impl.client.InternalHttpClient.doExecute(org.apache.http.HttpHost, org.apache.http.HttpRequest, org.apache.http.protocol.HttpContext) line: 185   
org.apache.http.impl.client.InternalHttpClient(org.apache.http.impl.client.CloseableHttpClient).execute(org.apache.http.client.methods.HttpUriRequest, org.apache.http.protocol.HttpContext) line: 83   
org.eclipse.ecf.provider.filetransfer.httpclient45.HttpClientFileSystemBrowser.runRequest() line: 246   
org.eclipse.ecf.provider.filetransfer.browse.AbstractFileSystemBrowser$DirectoryJob.run(org.eclipse.core.runtime.IProgressMonitor) line: 69   
org.eclipse.core.internal.jobs.Worker.run() line: 63   

There will be two additional retries, and if those fail too, the repository will fail to load (or the artifact will fail to download).   In my product catalog generator, I load a whole whack of repositories and this happens frequently enough that more often than not, one or more repositories fail to load.  This concerns me because with 2 million users loading repositories (via updates or via the installer), a significant fraction might well run into this same problem.

When I use -Dorg.eclipse.ecf.provider.filetransfer.excludeContributors=org.eclipse.ecf.provider.filetransfer.httpclient45 to disable the provider for this implementation, I of course see no log entries nor do any repositories fail to load.

I'm starting to get the feeling that either a) the server is ill-behaved or b) there is a problem in the httpclient implementation.  Perhaps some Keep-Alive header: is poor or not being respected by the implementation.

I saw something similar here where the server was blamed:

  https://support.sonatype.com/hc/en-us/articles/213465028-Understanding-The-target-server-failed-to-respond-in-Nexus-logs

Does anyone have any ideas?

Regards,
Ed





Back to the top