Fix elapsedSeconds date formatting causing netimid crash during BMC booting
If sending ipmi command during BMC booting process or BMC is very busy,
the netipmid service will crash.
The root cause is that the incoming UDP packet processing time > 2 seconds,
and triggers the log, but the elapsedSeconds date formatting is incorrect,
it causes the process crash.
Signed-off-by: Yong Li <yong.b.li@linux.intel.com>
Change-Id: Ia84149e4b4344793916003fc1ad0cdb26070e388
diff --git a/command_table.cpp b/command_table.cpp
index b29b32a..2a964ef 100644
--- a/command_table.cpp
+++ b/command_table.cpp
@@ -106,7 +106,7 @@
auto end = std::chrono::steady_clock::now();
- auto elapsedSeconds =
+ std::chrono::duration<size_t> elapsedSeconds =
std::chrono::duration_cast<std::chrono::seconds>(end - start);
// If command time execution time exceeds 2 seconds, log a time
@@ -114,7 +114,7 @@
if (elapsedSeconds > 2s)
{
log<level::ERR>("IPMI command timed out",
- entry("DELAY=%d", elapsedSeconds.count()));
+ entry("DELAY=%zu", elapsedSeconds.count()));
}
}
}