cURL / Mailing Lists / curl-library / Single Mail


Re: curl-library Digest, Vol 129, Issue 1

From: Ison <>
Date: Tue, 03 May 2016 19:08:15 +0530


/> echo nameserver > /etc/resolv.conf /
/> /
/> The file is getting updated and the operation exits with error
"Timeout was /
/> reached". /

Sorry, but is that the error message from curl when it fails?

/Yes that is the error message from curl when it fails. /

/> I am able to "ping" and "ping my_ftp_server" with as /
/> the only entry in the resolv.conf file, but the curl operation fails. /

Which libcurl version? On Linux? What name resolver backend does your
use? 'curl -V' shows.

/Sorry, I missed out the libcurl version in my querry. It is 7.45. I am
using debian jessie in Beaglebone Black.
I have cross compiled curl and is using the library from my application.
I have compiled the library with "--enable-threaded_resolver". curl -V
gives the following output:/

curl 7.48.0 (armv7l-unknown-linux-gnueabihf) libcurl/7.38.0
OpenSSL/1.0.1k zlib/1.2.8 libidn/1.29 libssh2/1.4.3 librtmp/2.3
Protocols: dict file ftp ftps gopher http https imap imaps ldap ldaps
pop3 pop3s rtmp rtsp scp sftp smtp smtps telnet tftp
libz TLS-SRP

Can you please be more specific about the exact error message and return
you get from libcurl?

/The error code is 28 and the error message is /Timeout was //reached//

As Dan Fandrich already asked you on curl-users: Can you use the Google DNS
server manually? e.g. does "host" or "dig @" return anything?

The following is the output for host

/root_at_arm:/home/debian# echo nameserver > /etc/resolv.conf
root_at_arm:/home/debian# cat /etc/resolv.conf
root_at_arm:/home/debian# host //
Using domain server:

// address // IPv6 address 2404:6800:4007:806::2004/


Ison Thomas

On Tuesday 03 May 2016 03:30 PM, wrote:
> Send curl-library mailing list submissions to
> To subscribe or unsubscribe via the World Wide Web, visit
> or, via email, send a message with subject or body 'help' to
> You can reach the person managing the list at
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of curl-library digest..."
> Today's Topics:
> 1. now checking more docs with tests (Daniel Stenberg)
> 2. Re: now checking more docs with tests (Dan Fandrich)
> 3. FTP operation timeout when public DNS Server is used (Ison)
> ----------------------------------------------------------------------
> Message: 1
> Date: Tue, 3 May 2016 00:01:11 +0200 (CEST)
> From: Daniel Stenberg <>
> To: libcurl hacking <>
> Subject: now checking more docs with tests
> Message-ID: <>
> Content-Type: text/plain; format=flowed; charset=US-ASCII
> Hey just a FYI here,
> Just wanted to mention that I added test 1139 and 1140 the last couple of
> days. They do all sorts of fun verifications of docs:
> test 1139 runs which
> 1. checks that all CURLOPT_ options mentioned in symbols-in-versions have
> their own man pages _and_ are mentioned in curl_easy_setopt.3
> 2. checks that all CURLINFO_ options mentioned in symbols-in-versions have
> their own man pages _and_ are mentioned in curl_easy_getinfo.3
> 3. checks that all CURLMOPT_ options mentioned in symbols-in-versions have
> their own man pages _and_ are mentioned in curl_multi_setopt.3
> 4. checks that the code that shows curl -h, the code that lists all command
> line options for the parser and curl.1 all mention the same set of command
> line options. Any and all mismatches are reported. There's a whitelisting
> concept to allow curl -h to not include deprecated options.
> test 1140 runs which goes through all nroff formatted man pages
> we have (about 346 files right now) and:
> 1. checks that all references to curl symbols are all within \fI and \fP
> 2. checks that the targets in references are actual existing man pages
> 3. checks for mentions of curl symbols with (3) added, that isn't written to
> become a reference
> While writing these scripts I of course detected a busload of mistakes in the
> docs that I've pushed. My hope is that these scripts will help us reduce the
> number of documentation mistakes we commit and help us detect them when we
> forget. As the tests also run on pull-requests, they will help contributors to
> get their patches with docs done correctly.
> Let me know if you get or see any problems because of these!

List admin:
Received on 2016-05-03