Ray Satiro via curl-library <curl-library@cool.haxx.se> wrote on 2021-01-10:

> On 1/10/2021 5:33 PM, Daniel Stenberg via curl-library wrote:
> > On Sun, 10 Jan 2021, Fabian Keil via curl-library wrote:
> >
> >> The attached patch seems to work around the problem.
> >> Busy-looping probably isn't the right solution, though.
> >
> > Yeah, that's just not the right fix.
> >
> > When the code returns there (http_proxy.c line 346) due to CURLE_AGAIN 
> > having been returned from the read function, it should return back out 
> > to the multi state-machine where it should remain in the 
> > CURLM_STATE_WAITPROXYCONNECT state until the socket gets readable 
> > again and it calls Curl_http_connect() again to continue to the 
> > CONNECT handling.
> >
> > Can you rebuild with --enable-debug and see if the extra output shines 
> > some more light on this? Presumably, in your case the code moves on to 
> > the next state somehow when it shouldn't and we should figure out why! 
> 
> 
> +1 we want it non-blocking. I can't reproduce this but I took a guess. 
> Can you please try https://github.com/curl/curl/pull/6432

Good guess. The commit 1253e5f4ba1210 makes the test
succeed 1000 times in a row.

Thanks.

Fabian

Attachment: pgpX9Ef_5l0oY.pgp
Description: OpenPGP digital signature

-------------------------------------------------------------------
Unsubscribe: https://cool.haxx.se/list/listinfo/curl-library
Etiquette:   https://curl.se/mail/etiquette.html

Reply via email to