Catch the correct sdbusplus exception
sdbusplus::exception::SdBusError is intended to be internal to the
sdbusplus library [1] and not depended on by client code:
['SdBusError'] was always intended to be internal to sdbusplus (in
fact, it inherits from an error called 'internal_error')
and `include/sdbusplus/exception.hpp`:
/** Exception for when an underlying sd_bus method call fails. */
class SdBusError final : public internal_exception
(Only) catching and handling `sdbusplus::exception::SdBusError` may lead
to some sdbusplus exceptions being unhandled and thus propagating [1]:
Quite likely, by catching this exception, you are missing other
exceptions that sdbusplus can throw.
The correct general exception seems to be
`sdbusplus::exception::exception`.
Tested: Executed the affected `asio-example` with and without this
change and verified the output was the same (except for differences in
the order of printed messages, which seems to vary from invocation to
another also without the change).
[1] The message "sdbusplus exception type SdBusError" on the sdbusplus mailing list: https://lore.kernel.org/openbmc/YTDvfIn4Z05mGdCx@heinlein/
Change-Id: I5cda83c593730f6c0ce5735fb7c46d43f2066c9d
Signed-off-by: Hannu Lounento <hannu.lounento@vaisala.com>
diff --git a/example/asio-example.cpp b/example/asio-example.cpp
index 906059e..8138dea 100644
--- a/example/asio-example.cpp
+++ b/example/asio-example.cpp
@@ -114,7 +114,7 @@
{
reply.read(tupleOut);
}
- catch (const sdbusplus::exception::SdBusError& e)
+ catch (const sdbusplus::exception::exception& e)
{
std::cerr << "failed to unpack; sig is " << reply.get_signature()
<< "\n";
@@ -288,7 +288,7 @@
message intMsg = conn->call(readyMsg);
intMsg.read(ready);
}
- catch (const sdbusplus::exception::SdBusError& e)
+ catch (const sdbusplus::exception::exception& e)
{
ready = 0;
// pause to give the server a chance to start up