DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Robin Jarry" <rjarry@redhat.com>
To: "Bruce Richardson" <bruce.richardson@intel.com>
Cc: <dev@dpdk.org>, <stable@dpdk.org>
Subject: Re: [PATCH dpdk] telemetry-exporter: listen on loopback by default
Date: Mon, 27 Jan 2025 21:21:34 +0100	[thread overview]
Message-ID: <D7D4VJ9GISNJ.2RIGH6IDPR8BL@redhat.com> (raw)
In-Reply-To: <Z5eauLXHdzkNi64w@bricha3-mobl1.ger.corp.intel.com>

Bruce Richardson, Jan 27, 2025 at 15:39:
> On Mon, Jan 27, 2025 at 12:51:44PM +0100, Robin Jarry wrote:
>> Fix the following warning reported by Coverity:
>> 
>> Defect type: SIGMA.insecure_network_bind:
>> > dpdk-stable-24.11.1/usertools/dpdk-telemetry-exporter.py:278:
>> > Sigma main event: The HTTP server binds to all network interfaces by
>> > setting the IP address to "", `0.0.0.0`, `::`, or `::0`.
>> > This may expose the server to unintended traffic.
>> 
>> Avoid listening to all interfaces by default to avoid exposing private
>> information unwillingly.
>> 
>> Unrelated: The Python stdlib TCP server listens on IPv4 only by default.
>> Changing this requires creating a subclass that overrides address_family
>> to socket.AF_INET6.
>> 
>> Fixes: d94ebd627a86 ("usertools: add telemetry exporter")
>> Cc: stable@dpdk.org
>> Signed-off-by: Robin Jarry <rjarry@redhat.com>
>> ---
>>  usertools/dpdk-telemetry-exporter.py | 6 +++---
>>  1 file changed, 3 insertions(+), 3 deletions(-)
>> 
>> diff --git a/usertools/dpdk-telemetry-exporter.py b/usertools/dpdk-telemetry-exporter.py
>> index 6eca0db2e80a..6f66d4ecaab1 100755
>> --- a/usertools/dpdk-telemetry-exporter.py
>> +++ b/usertools/dpdk-telemetry-exporter.py
>> @@ -75,7 +75,7 @@ def cmd(self, uri, arg=None) -> dict | list:
>>      "/usr/local/share/dpdk/telemetry-endpoints",
>>      "/usr/share/dpdk/telemetry-endpoints",
>>  ]
>> -DEFAULT_OUTPUT = "openmetrics://:9876"
>> +DEFAULT_OUTPUT = "openmetrics://127.0.0.1:9876"
>>  
>
> Minor nit, but would it be better to use "localhost" rather than the
> hardcoded IP here and below?

That's a good point. I had considered it but as I wrote in the commit
message:

>> Unrelated: The Python stdlib TCP server listens on IPv4 only by default.
>> Changing this requires creating a subclass that overrides address_family
>> to socket.AF_INET6.

On certain systems and depending on the libc implementation, localhost
may resolve to ::1 which causes an error on startup:

  socket.gaierror: [Errno -9] Address family for hostname not supported

I dug a bit and it happens that the python standard TCPServer
implementation explicitly uses AF_INET when creating the socket. Hence
bind() fails with IPv6 addresses.


      reply	other threads:[~2025-01-27 20:21 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-27 11:51 Robin Jarry
2025-01-27 14:39 ` Bruce Richardson
2025-01-27 20:21   ` Robin Jarry [this message]

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=D7D4VJ9GISNJ.2RIGH6IDPR8BL@redhat.com \
    --to=rjarry@redhat.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=stable@dpdk.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).