From: "v.shpilevoy@tarantool.org" <v.shpilevoy@tarantool.org>
To: tarantool-patches@freelists.org
Cc: Konstantin Osipov <kostja@tarantool.org>, vdavydov.dev@gmail.com
Subject: Re: [tarantool-patches] [PATCH 1/3] netbox: allow to create a netbox connection from existing socket
Date: Fri, 23 Mar 2018 13:01:45 +0300 [thread overview]
Message-ID: <E2C340B8-6734-4EB0-B0A4-4A1227217A77@tarantool.org> (raw)
In-Reply-To: <9D9CC51C-2CFA-4B34-BDC6-2E82A7724E16@tarantool.org>
I have found, that SIGPIPE appears only under a debugger (lldb or gdb), but it does not mean, that it's ok to crash during debugging. I leave the try-to-read-before-write fix, but I made it slightly more simple. Instead of storing read symbols in a buffer I just mark them as unread in ibuf. The patch is below and on the branch.
diff --git a/src/box/lua/console.lua b/src/box/lua/console.lua
index 02e7f1066..633144c7f 100644
--- a/src/box/lua/console.lua
+++ b/src/box/lua/console.lua
@@ -94,20 +94,19 @@ local text_connection_mt = {
-- @retval nil Error.
--
write = function(self, text)
- -- It is the hack to protect from SIGPIPE on send in
+ -- It is the hack to protect from SIGPIPE, which is
+ -- not ignored under debugger (gdb, lldb) on send in
-- a socket, that is actually closed. If a socket is
-- readable and read() returns nothing then the socket
-- is closed, and writing into it will raise SIGPIPE.
if self._socket:readable(0) then
- local rc = self._socket:recv(1)
+ local rc = self._socket:read({chunk = 1})
if not rc or rc == '' then
return nil
else
- -- But if it was literally readable, the
- -- single read byte can not be put back, and
- -- must be stored somewhere, until console
- -- do read().
- self.buffer = self.buffer..rc
+ assert(#rc == 1)
+ -- Make the char be unread.
+ self._socket.rbuf.wpos = self._socket.rbuf.wpos - 1
end
end
return self._socket:send(text)
@@ -120,8 +119,6 @@ local text_connection_mt = {
read = function(self)
local ret = self._socket:read(YAML_TERM)
if ret and ret ~= '' then
- ret = (self.buffer or '')..ret
- self.buffer = nil
return ret
end
end,
@@ -168,7 +165,6 @@ local function wrap_text_socket(connection, url)
state = 'active',
host = url.host or 'localhost',
port = url.service,
- buffer = nil,
}, text_connection_mt)
if not conn:write('require("console").delimiter("$EOF$")\n') or
not conn:read() then
next prev parent reply other threads:[~2018-03-23 10:01 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-22 19:17 [PATCH 0/3] console: do not use netbox for console text connections Vladislav Shpilevoy
2018-03-22 19:17 ` [PATCH 1/3] netbox: allow to create a netbox connection from existing socket Vladislav Shpilevoy
2018-03-22 19:33 ` [tarantool-patches] " Konstantin Osipov
2018-03-22 19:50 ` v.shpilevoy
2018-03-22 19:57 ` [tarantool-patches] " v.shpilevoy
2018-03-23 10:01 ` v.shpilevoy [this message]
2018-03-26 21:55 ` [tarantool-patches] " Konstantin Osipov
2018-03-22 19:17 ` [PATCH 2/3] console: do not use netbox for console text connections Vladislav Shpilevoy
2018-03-22 19:36 ` [tarantool-patches] " Konstantin Osipov
2018-03-22 19:17 ` [PATCH 3/3] netbox: deprecate console support Vladislav Shpilevoy
2018-03-22 19:37 ` [tarantool-patches] " Konstantin Osipov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=E2C340B8-6734-4EB0-B0A4-4A1227217A77@tarantool.org \
--to=v.shpilevoy@tarantool.org \
--cc=kostja@tarantool.org \
--cc=tarantool-patches@freelists.org \
--cc=vdavydov.dev@gmail.com \
--subject='Re: [tarantool-patches] [PATCH 1/3] netbox: allow to create a netbox connection from existing socket' \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox