fdo-mirrors/NetworkManager
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/NetworkManager/NetworkManager.git synced 2026-05-05 20:28:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

core: don't persist secret-key for tests

Browse source 
Tests might access the secret-key.

For CI builds we may very well build NM as root and also run
unit tests. In such a situation it's bad to persist the secret
key. For example, the SELinux label may be wrong, and subsequently
starting NetworkManager may cause errors. Avoid persisting the secret
key for tests.
This commit is contained in:
Thomas Haller 2018-11-13 18:50:01 +01:00
parent 8308311264
commit 581e1c3269
2 changed files with 8 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
src/nm-core-utils.c
View file

@ -2511,6 +2511,12 @@ _secret_key_read (guint8 **out_secret_key,
goto out;
}
if (nm_utils_get_testing ()) {
/* for test code, we don't write the generated secret-key to disk. */
success = FALSE;
goto out;
}
if (!nm_utils_file_set_contents (NMSTATEDIR "/secret_key", (char *) secret_key, key_len, 0077, &error)) {
nm_log_warn (LOGD_CORE, "secret-key: failure to persist secret key in \"%s\" (%s) (use non-persistent key)",
NMSTATEDIR "/secret_key", error->message);

8
src/tests/test-general.c
View file

@ -1922,12 +1922,8 @@ test_machine_id_read (void)
logstate = nmtst_logging_disable (FALSE);
/* If you run this test as root, without a valid /etc/machine-id,
* the code will try to get the secret-key (and possibly attempt
* to write it).
*
* That's especially ugly, if you run the test as root and it writes
* a new "/var/lib/NetworkManager/secret_key" file. Another reason
* not to run tests as root. */
* the code will try to get the secret-key. That is a bit ugly,
* but no real problem. */
machine_id = nm_utils_machine_id_bin ();
nmtst_logging_reenable (logstate);