summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* RSS aware maxmemory: algorith #1 implemented.rssmaxmemoryantirez2014-11-173-15/+68
|
* RSS aware maxmemory: enforced/adjusted == maxmemory on config changes.antirez2014-11-172-2/+6
|
* RSS aware maxmemory: config settings.antirez2014-11-174-0/+25
|
* THP detection for LATENCY DOCTOR.antirez2014-11-121-2/+13
|
* Check THP support at startup and warn about it.antirez2014-11-122-2/+7
|
* THP detection / reporting functions added.antirez2014-11-123-6/+45
|
* Diskless SYNC: fix RDB EOF detection.antirez2014-11-113-4/+21
| | | | | | | | | | | | | | | | | | | RDB EOF detection was relying on the final part of the RDB transfer to be a magic 40 bytes EOF marker. However as the slave is put online immediately, and because of sockets timeouts, the replication stream is actually contiguous with the RDB file. This means that to detect the EOF correctly we should either: 1) Scan all the stream searching for the mark. Sucks CPU-wise. 2) Start to send the replication stream only after an acknowledge. 3) Implement a proper chunked encoding. For now solution "2" was picked, so the master does not start to send ASAP the stream of commands in the case of diskless replication. We wait for the first REPLCONF ACK command from the slave, that certifies us that the slave correctly loaded the RDB file and is ready to get more data.
* Disconnect timedout slave: regression introduced with diskless repl.antirez2014-11-111-2/+3
|
* Merge pull request #2096 from mattsta/cluster-ipv6Salvatore Sanfilippo2014-10-312-6/+8
|\ | | | | Enable Cluster IPv6 Support
| * Parse cluster state file in IPv6 compatible wayMatt Stancliff2014-10-291-1/+1
| | | | | | | | We need to pick the port based on the _last_ colon, not the first one.
| * Fix redis-trib.rb IP:Port disassembly for IPv6Matt Stancliff2014-10-291-5/+7
| | | | | | | | | | | | IP format is now any of: - 127.0.0.1:6379 - ::1:6379
* | Merge pull request #2110 from mattsta/more-outbound-bind-fixesSalvatore Sanfilippo2014-10-312-3/+4
|\ \ | | | | | | Networking: add more outbound IP binding fixes
| * | Networking: add more outbound IP binding fixesMatt Stancliff2014-10-292-3/+4
| |/ | | | | | | | | | | | | | | | | | | | | Same as the original bind fixes (we just missed these the first time around). This helps Redis not automatically send connections from the first IP on an interface if we are bound to a specific IP address (e.g. with multiple IP aliases on one interface, you want to send from _your_ IP, not from the first IP on the interface).
* | redis-cli: ignore SIGPIPE network errorsMatt Stancliff2014-10-291-1/+4
|/ | | | Closes #2066
* Merge branch 'memsync' into unstableantirez2014-10-2912-134/+859
|\
| * Diskless replication: missing listRewind() added.memsyncantirez2014-10-291-1/+5
| | | | | | | | | | | | | | | | | | | | This caused BGSAVE to be triggered a second time without any need when we switch from socket to disk target via the command CONFIG SET repl-diskless-sync no and there is already a slave waiting for the BGSAVE to start. Also comments clarified about what is happening.
| * Log slave ip:port in more log messages.antirez2014-10-271-6/+11
| |
| * Use new slave name function for diskless repl reporting.antirez2014-10-271-4/+4
| |
| * Added a function to get slave name for logs.antirez2014-10-273-15/+29
| |
| * Diskless replication: log BGSAVE delay only when it is non-zero.antirez2014-10-271-1/+2
| |
| * Diskless sync delay is now configurable.antirez2014-10-274-3/+17
| |
| * Remove duplicated log message about starting BGSAVE.antirez2014-10-241-1/+0
| |
| * Diskless replication: child -> parent communication improved.antirez2014-10-231-20/+34
| | | | | | | | | | Child now reports full info to the parent including IDs of slaves in failure state and exit code.
| * Translate rio fdset target EWOULDBLOCK error into ETIMEDOUT.antirez2014-10-221-1/+8
| | | | | | | | | | | | | | EWOULDBLOCK with the fdset rio target is returned when we try to write but the send timeout socket option triggered an error. Better to translate the error in something the user can actually recognize as a timeout.
| * Diskless replication: set / reset socket send timeout.antirez2014-10-221-0/+2
| | | | | | | | | | | | We need to avoid that a child -> slaves transfer can continue forever. We use the same timeout used as global replication timeout, which is documented to also affect I/O operations during bulk transfers.
| * anet.c: new API anetSendTimeout().antirez2014-10-222-0/+16
| |
| * Diskless replication: less debugging printfs around.antirez2014-10-171-1/+0
| |
| * rio.c fdset write() method fixed: wrong type for return value.antirez2014-10-171-1/+1
| |
| * rio fdset target: handle short writes.antirez2014-10-173-2/+18
| | | | | | | | | | While the socket is set in blocking mode, we still can get short writes writing to a socket.
| * anet.c: API to set sockets back to blocking mode.antirez2014-10-172-4/+18
| |
| * Diskless replication: rio fdset target new supports buffering.antirez2014-10-173-1/+56
| | | | | | | | | | | | | | | | | | | | | | | | | | | | To perform a socket write() for each RDB rio API write call was extremely unefficient, so now rio has minimal buffering capabilities. Writes are accumulated into a buffer and only when a given limit is reacehd are actually wrote to the N slaves FDs. Trivia: rio lacked support for buffering since our targets were: 1) Memory buffers. 2) C standard I/O. Both were buffered already.
| * Diskless replication: Various fixes to backgroundSaveDoneHandlerSocket()antirez2014-10-171-4/+8
| |
| * Diskless replication: don't send "\n" pings to slaves.antirez2014-10-171-1/+3
| | | | | | | | | | | | This is useful for normal replication in order to refresh the slave when we are persisting on disk, but for diskless replication the child is already receiving data while in WAIT_BGSAVE_END state.
| * Diskless replication: remove 40 bytes EOF mark from end of RDB file.antirez2014-10-171-0/+10
| |
| * Diskless replication: swap inverted branches to compute read len.antirez2014-10-171-2/+2
| |
| * Diskless replication: don't enter the read-payload branch forever.antirez2014-10-171-0/+3
| |
| * syncReadLine(): actually enforce buffer size limits.antirez2014-10-161-0/+1
| |
| * Diskless replication: EOF:<mark> streaming support slave side.antirez2014-10-161-7/+60
| |
| * Diskless replication: redis.conf and CONFIG SET/GET support.antirez2014-10-164-5/+18
| |
| * Diskless replication: trigger a BGSAVE after a config change.antirez2014-10-161-12/+17
| | | | | | | | | | | | | | | | | | If we turn from diskless to disk-based replication via CONFIG SET, we need a way to start a BGSAVE if there are slaves alerady waiting for a BGSAVE to start. Normally with disk-based replication we do it as soon as the previous child exits, but when there is a configuration change via CONFIG SET, we may have slaves in WAIT_BGSAVE_START state without an RDB background process currently active.
| * Diskless replication flag renamed repl_diskless -> repl_diskless_sync.antirez2014-10-163-9/+9
| |
| * Diskless replication: trigger diskless RDB transfer if needed.antirez2014-10-162-2/+37
| |
| * Diskless replication: handle putting the slave online.antirez2014-10-151-31/+57
| |
| * Diskless replication: read report from child.antirez2014-10-151-15/+57
| |
| * Diskless replication: child writes report to parent.antirez2014-10-151-0/+40
| |
| * rio.c fdset target: tolerate (and report) a subset of FDs in error.antirez2014-10-142-3/+22
| | | | | | | | | | | | | | | | | | | | | | | | Fdset target is used when we want to write an RDB file directly to slave's sockets. In this setup as long as there is a single slave that is still receiving our payload, we want to continue sennding instead of aborting. However rio calls should abort of no FD is ok. Also we want the errors reported so that we can signal the parent who is ok and who is broken, so there is a new set integers with the state of each fd. Zero is ok, non-zero is the errno of the failure, if avaialble, or a generic EIO.
| * Diskless replication: parent-child pipe and a few TODOs.antirez2014-10-142-4/+35
| |
| * Diskless replication: RDB -> slaves transfer draft implementation.antirez2014-10-145-28/+200
| |
| * rio.c: draft implementation of fdset target implemented.antirez2014-10-102-0/+69
| |
| * rio.c refactoring before adding a new target.antirez2014-10-101-17/+24
| |
View Lorry Controller Status