Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

8.3: Import netdata v2.1.0 #5

Open
wants to merge 18 commits into
base: master
Choose a base branch
from

Conversation

tescande
Copy link
Contributor

No description provided.

Signed-off-by: Thierry Escande <[email protected]>
@tescande tescande force-pushed the for-8.3/import_netdata-2.1.0-3.el9.src.rpm branch from 36913d7 to 6863e5e Compare January 20, 2025 12:38
@tescande tescande marked this pull request as draft January 20, 2025 17:26
tescande and others added 16 commits January 20, 2025 19:16
This merge gets rid of the previous changes made on top of netdata v1.44
and import netdata v2.1 from netdata-2.1.0-3.el9.src.rpm. All needed
modifications will be ported on top of this merge point.

Signed-off-by: Thierry Escande <[email protected]>
This patch fixes compiler errors "initializer element is not constant"
in global static structures initialization.

The defines for default struct fields are defined as:

 #define DEFAULT_FOO (struct foo){.bar = 1, ...}

And then used in global static structure declarations as:

 static struct {
     int a;
     int b;
     struct foo f;
     ....;
 } toz = {
     .a = 1,
     .b = 2,
     .f = DEFAULT_FOO,
     ....,
 }

Our gcc 4.8 doesn't like the type cast in DEFAULT_FOO definition,
leading to the compilation error. So this patch removes the typecast in
the default values definition.

But then, in-code assignation fails for lack of typecast:

struct foo f = DEFAULT_FOO;

Generating the error "expected expression before '{' token".
This patch adds the correct typecast in such cases.

Signed-off-by: Thierry Escande <[email protected]>
Signed-off-by: Thierry Escande <[email protected]>
Signed-off-by: Thierry Escande <[email protected]>
I'm all for sending anonymous usage data, but I prefer our users to
opt-in rather than opt-out.

Signed-off-by: Samuel Verschelde <[email protected]>
Signed-off-by: Thierry Escande <[email protected]>
- Expose the netdata port
- Explicitly restrict access to netdata.conf to localhost
- Use db ram mode instead of dbengine to avoid disk flooding

Signed-off-by: Thierry Escande <[email protected]>
- The netdata package comes with a configuration with web UI disabled
- The netdata-ui package enables the web UI.

Signed-off-by: Samuel Verschelde <[email protected]>
Signed-off-by: Thierry Escande <[email protected]>
Installing netdata-ui will add firewall rules to open port 19999
when the service starts. Users can modify this behaviour by changing the
contents of /etc/sysconfig/iptables_netdata and
/etc/sysconfig/ip6tables_netdata.

One can even create empty files with those names before installing the
netdata package to prevent the port from being automatically open.

Installing the netdata package alone does not enable the web UI and does
not open the 19999 port.

Signed-off-by: Samuel Verschelde <[email protected]>
Signed-off-by: Thierry Escande <[email protected]>
Signed-off-by: Thierry Escande <[email protected]>
When removing netdata and netdata-ui packages together (same yum
transaction), the netdata service is restarted by the ui package then
stopped immediately after by the removal the netdata package. This could
make the service to hang and leave it in a failed state and not entirely
removed. This patch adds a delay of 3 seconds before killing the main
netdata process. This gives enough time to netdata to start before being
stopped.

Signed-off-by: Thierry Escande <[email protected]>
With this change, plugin files don't get cap_setuid=pe since the setuid
bit is already set in their attributes. This avoid a warning in kernel
dmesg regarding files having both cap_setuid anf the setuid bit set.

Signed-off-by: Thierry Escande <[email protected]>
@tescande tescande force-pushed the for-8.3/import_netdata-2.1.0-3.el9.src.rpm branch from 96e7d53 to 8351353 Compare January 20, 2025 18:18
@tescande tescande marked this pull request as ready for review January 20, 2025 18:18
Signed-off-by: Thierry Escande <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants