Difference between revisions of "Sockets"
(Created page with "== Unix sockets initial support == Currently it can only work with stream sockets, which have no skbs in queues (listening or established -- both work OK). The cpt part uses th...") |
|||
| Line 2: | Line 2: | ||
Currently it can only work with stream sockets, which have no skbs in queues | Currently it can only work with stream sockets, which have no skbs in queues | ||
| − | (listening or established -- both work OK). | + | (listening or established -- both work OK). If there are skbs in queue the dumping |
| + | procedure will fail. | ||
The cpt part uses the sock_diag engine that was merged to Dave recently to | The cpt part uses the sock_diag engine that was merged to Dave recently to | ||
| Line 9: | Line 10: | ||
such tricks with opens through proc) against SOCKFS_TYPE. | such tricks with opens through proc) against SOCKFS_TYPE. | ||
| − | The | + | The restore part is more tricky. Listen sockets are just restored, this is simple. |
Connected sockets are restored like this: | Connected sockets are restored like this: | ||
| Line 34: | Line 35: | ||
* Implement support for UDP sockets (quite simple) | * Implement support for UDP sockets (quite simple) | ||
* Implement support for listening TCP sockets (also not very complex) | * Implement support for listening TCP sockets (also not very complex) | ||
| − | * Implement support for connected TCP scokets (hard one | + | * Implement support for connected TCP scokets (hard one) |
Revision as of 19:44, 27 December 2011
Unix sockets initial support
Currently it can only work with stream sockets, which have no skbs in queues (listening or established -- both work OK). If there are skbs in queue the dumping procedure will fail.
The cpt part uses the sock_diag engine that was merged to Dave recently to collect sockets. Then it dumps sockets by checking the filesystem ID of a failed-to-open through /proc/pid/fd descriptors (sockets do not allow for such tricks with opens through proc) against SOCKFS_TYPE.
The restore part is more tricky. Listen sockets are just restored, this is simple. Connected sockets are restored like this:
- One end establishes a listening anon socket at the desired descriptor;
- The other end just creates a socket at the desired descriptor;
- All sockets, that are to be connect()-ed call connect. Unix sockets do not block connect() till the accept() time and thus we continue with...
- ... all listening sockets call accept() and ... dup2 the new fd into the accepting end.
There's a problem with this approach -- socket names are not preserved, but looking into our OpenVZ implementation I think this is OK for existing apps.
What should be done next
- Need to merge the file IDs patches in our tree and make Andrey to support files sharing. This will solve the
sk = socket();
fork();
case. Currently it simply doesn't work :(
- Need to add support for DGRAM sockets -- I wrote comment how to do it in the can_dump_unix_sk()
- Need to add support for in-flight connections
- Implement support for UDP sockets (quite simple)
- Implement support for listening TCP sockets (also not very complex)
- Implement support for connected TCP scokets (hard one)