Conversation
| if ((status == ARES_ENODATA || status == ARES_EBADRESP) && | ||
| hquery->want_family == AF_UNSPEC) { | ||
| if ((status == ARES_ENODATA || status == ARES_EBADRESP || | ||
| (status == ARES_SUCCESS && host && host->h_addr_list[0] == NULL)) && |
There was a problem hiding this comment.
Just recalling, are we going to switch io.js to use the proper error names instead of the made up one we have (ENOTFOUND)?
There was a problem hiding this comment.
I have no love for ENOTFOUND but renaming it may result in a lot of breakage. :-/
|
Change looks fine to me. |
|
LGTM if CI is happy. |
Fixes: nodejs#1676 PR-URL: nodejs#1678 Reviewed-By: Fedor Indutny <fedor@indutny.com> Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Provide more information in `ares_txt_reply` to coalesce chunks from the same record into one string. fix nodejs#7367
On Windows, when compiling with `UNICODE` defined, `LoadLibrary` becomes `LoadLibraryW`. When an ASCII string is passed to that function it crashes. PR-URL: nodejs#226 Reviewed-By: Bert Belder <bertbelder@gmail.com>
|
Landed in 7e1c0e7...08d0866. Thanks for the review, guys. There are some CI failures but not ones introduced by this PR, AFAICT. (Child process failures on Windows, a dgram test that has been failing for some time now. There are issues about them.) I decided to put not too much stock in the test failure on the armv7-wheezy buildbot because it always has one or two tests randomly failing (sadly.) If the failure turns out to be persistent, I'm on the hook for fixing it. |
Fixes: nodejs#1676 PR-URL: nodejs#1678 Reviewed-By: Fedor Indutny <fedor@indutny.com> Reviewed-By: Trevor Norris <trev.norris@gmail.com>
R=@indutny?
CI: https://jenkins-iojs.nodesource.com/view/iojs/job/iojs+any-pr+multi/661/