Should be much faster, and regardless, the code is more obvious now
compared to generating a string just to get the return value of the
ll2stirng() function.
1. HVSTRLEN -> HSTRLEN. It's unlikely one needs the length of the key,
not clear how the API would work (by value does not make sense) and
there will be better names anyway.
2. Default is to return 0 when field is missing.
3. Default is to return 0 when key is missing.
4. The implementation was slower than needed, and produced unnecessary COW.
Related issue #2415.
1. Remove useless "cs" initialization.
2. Add a "select" var to capture a condition checked multiple times.
3. Avoid duplication of the same if (!copy) conditional.
4. Don't increment dirty if copy is given (no deletion is performed),
otherwise we propagate MIGRATE when not needed.
Less grays: more readable palette since usually we have a non linear
distribution of percentages and very near gray tones are hard to take
apart. Final part of the palette is gradient from yellow to red. The red
part is hardly reached because of usual distribution of latencies, but
shows up mainly when latencies are very high because of the logarithmic
scale, this is coherent to what people expect: red = bad.
The old version of SPOP with "count" argument used an API call of dict.c
which was actually designed for a different goal, and was not capable of
good distribution. We follow a different three-cases approach optimized
for different ratiion between sets and requested number of elements.
The implementation is simpler and allowed the removal of a large amount
of code.
Severan problems are addressed but still a few missing.
Since replication of this command was more complex than others since it
needs to replicate multiple SREM commands, an old API able to do this
was reused (it was taken inside the implementation since it was pretty
obvious soon or later that would be useful). The API was improved a bit
so that now a command may opt-out for the standard command replication
when the server.dirty counter is incremented, in order to "manually"
replicate what it wants.
--stat mode already used to reconnect automatically if the server is no
longer available. This is useful since this is an interactive mode used
for debugging, however the same applies to --latency and --latency-dist
modes, so now both use the reconnecting command execution as well.
The reconnection code was modified to use basic VT100 escape sequences
in order to play better with different kinds of output on the screen
when the reconnection happens, and to hide the reconnection attempt
output when finally the reconnection happens.
So far not able to find a color palette within the 256 colors which is
not confusing. However I believe it is a possible task, so will try
better later.
This also makes it backward compatible in the usage, but for the command
name. However the old command name was less obvious so it is worth to
break it probably.
With the new setup the program main can perform argument parsing and
everything else useful for an RDB check regardless of the Redis server
itself.
When trying to debug sentinel connections or max connections errors it
would be very useful to have the ability to see the list of connected
clients to a running sentinel. At the same time it would be very helpful
to be able to name each sentinel connection or kill offending clients.
This commits adds the already defined CLIENT commands back to Redis
Sentinel.
This improves PFAIL -> FAIL switch. Too late at this point in the RC
releases to add proper PFAIL/FAIL separate dictionary to do this in a
less randomized way. Tested in practice with experiments that this
helps. PFAIL -> FAIL average with 20 nodes and node-timeout set to 5
seconds takes 2.5 seconds without this commit, 1 second with this
commit.
Otherwise it is impossible to receive the majority of failure reports in
the node_timeout*2 window in larger clusters.
Still with a 200 nodes cluster, 20 gossip sections are a very reasonable
amount of bytes to send.
A side effect of this change is also fater cluster nodes joins for large
clusters, because the cluster layout makes less time to propagate.
Previouly if we loaded a corrupt RDB, Redis printed an error report
with a big "REPORT ON GITHUB" message at the bottom. But, we know
RDB load failures are corrupt data, not corrupt code.
Now when RDB failure is detected (duplicate keys or unknown data
types in the file), we run check-rdb against the RDB then exit. The
automatic check-rdb hopefully gives the user instant feedback
about what is wrong instead of providing a mysterious stack
trace.
redis-check-rdb (previously redis-check-dump) had every RDB define
copy/pasted from rdb.h and some defines copied from redis.h. Since
the initial copy, some constants had changed in Redis headers and
check-dump was using incorrect values.
Since check-rdb is now a mode of Redis, the old check-dump code
is cleaned up to:
- replace all printf with redisLog (and remove \n from all strings)
- remove all copy/pasted defines to use defines from rdb.h and redis.h
- replace all malloc/free with zmalloc/zfree
- remove unnecessary include headers
redis-check-dump is now named redis-check-rdb and it runs
as a mode of redis-server instead of an independent binary.
You can now use 'redis-server redis.conf --check-rdb' to check
the RDB defined in redis.conf. Using argument --check-rdb
checks the RDB and exits. We could potentially also allow
the server to continue starting if the RDB check succeeds.
This change also enables us to use RDB checking programatically
from inside Redis for certain failure conditions.
Otherwise we risk sending not initialized data to other nodes, that may
contain anything. This was actually not possible only because the
initialization of the buffer where the cluster packets header is created
was larger than the 3 gossip sections we use, so the memory was already
all filled with zeroes by the memset().
On Darwin /dev/urandom depletes terribly fast. This is not an issue
normally, but with Redis Cluster we generate a lot of unique IDs, for
example during nodes handshakes. Our IDs need just to be unique without
other strong crypto requirements, so this commit turns the function into
something that gets a 20 bytes seed from /dev/urandom, and produces the
rest of the output just using SHA1 in counter mode.
Fixes valgrind error:
48 bytes in 1 blocks are definitely lost in loss record 196 of 373
at 0x4910D3: je_malloc (jemalloc.c:944)
by 0x42807D: zmalloc (zmalloc.c:125)
by 0x41FA0D: dictGetIterator (dict.c:543)
by 0x41FA48: dictGetSafeIterator (dict.c:555)
by 0x459B73: clusterHandleSlaveMigration (cluster.c:2776)
by 0x45BF27: clusterCron (cluster.c:3123)
by 0x423344: serverCron (redis.c:1239)
by 0x41D6CD: aeProcessEvents (ae.c:311)
by 0x41D8EA: aeMain (ae.c:455)
by 0x41A84B: main (redis.c:3832)
If array has N elements, we can't read +1 if we are already at N.
Also, we need to move elements by their storage size in the array,
not just by individual bytes.
[maybe] Fixes valgrind errors:
32 bytes in 4 blocks are definitely lost in loss record 107 of 228
at 0x80EA447: je_malloc (jemalloc.c:944)
by 0x806E59C: zrealloc (zmalloc.c:125)
by 0x80A9AFC: clusterSetMaster (cluster.c:801)
by 0x80AEDC9: clusterCommand (cluster.c:3994)
by 0x80682A5: call (redis.c:2049)
by 0x8068A20: processCommand (redis.c:2309)
by 0x8076497: processInputBuffer (networking.c:1143)
by 0x8073BAF: readQueryFromClient (networking.c:1208)
by 0x8060E98: aeProcessEvents (ae.c:412)
by 0x806123B: aeMain (ae.c:455)
by 0x806C3DB: main (redis.c:3832)
64 bytes in 8 blocks are definitely lost in loss record 143 of 228
at 0x80EA447: je_malloc (jemalloc.c:944)
by 0x806E59C: zrealloc (zmalloc.c:125)
by 0x80AAB40: clusterProcessPacket (cluster.c:801)
by 0x80A847F: clusterReadHandler (cluster.c:1975)
by 0x30000FF: ???
80 bytes in 10 blocks are definitely lost in loss record 148 of 228
at 0x80EA447: je_malloc (jemalloc.c:944)
by 0x806E59C: zrealloc (zmalloc.c:125)
by 0x80AAB40: clusterProcessPacket (cluster.c:801)
by 0x80A847F: clusterReadHandler (cluster.c:1975)
by 0x2FFFFFF: ???
Fixes valgrind error:
Syscall param write(buf) points to uninitialised byte(s)
at 0x514C35D: ??? (syscall-template.S:81)
by 0x456B81: clusterWriteHandler (cluster.c:1907)
by 0x41D596: aeProcessEvents (ae.c:416)
by 0x41D8EA: aeMain (ae.c:455)
by 0x41A84B: main (redis.c:3832)
Address 0x5f268e2 is 2,274 bytes inside a block of size 8,192 alloc'd
at 0x4932D1: je_realloc (jemalloc.c:1297)
by 0x428185: zrealloc (zmalloc.c:162)
by 0x4269E0: sdsMakeRoomFor.part.0 (sds.c:142)
by 0x426CD7: sdscatlen (sds.c:251)
by 0x4579E7: clusterSendMessage (cluster.c:1995)
by 0x45805A: clusterSendPing (cluster.c:2140)
by 0x45BB03: clusterCron (cluster.c:2944)
by 0x423344: serverCron (redis.c:1239)
by 0x41D6CD: aeProcessEvents (ae.c:311)
by 0x41D8EA: aeMain (ae.c:455)
by 0x41A84B: main (redis.c:3832)
Uninitialised value was created by a stack allocation
at 0x457810: nodeUpdateAddressIfNeeded (cluster.c:1236)
The cleanup code expects that if 'di' is not NULL, it is a valid
iterator that should be freed.
The result of this bug was a crash of the AOF rewriting process if an
error occurred after the DBs data are written and the iterator is no
longer valid.
Rationale is that when re-entering, it is likely due to Lua debugging
hooks. Returning an error will be ignored in most cases, going totally
unnoticed. With the log at least we leave a trace.
Related to issue #2302.
read() and write() return ssize_t (signed long), not int.
For other offsets, we can use the unsigned size_t type instead
of a signed offset (since our replication offsets and buffer
positions are never negative).
It's possible large objects could be larger than 'int', so let's
upgrade all size counters to ssize_t.
This also fixes rdbSaveObject serialized bytes calculation.
Since entire serializations of data structures can be large,
so we don't want to limit their calculated size to a 32 bit signed max.
This commit increases object size calculation and
cascades the change back up to serializedlength printing.
Before:
127.0.0.1:6379> debug object hihihi
... encoding:quicklist serializedlength:-2147483559 ...
After:
127.0.0.1:6379> debug object hihihi
... encoding:quicklist serializedlength:2147483737 ...
In order to avoid that misconfigured cluster nodes at some time may
force an IP update on other nodes, it is required that nodes update
their own address only on MEET messages. However it does not make sense
to do this the first time a node is contacted and yet does not have an
IP, we just risk that myself->ip remains not assigned if there are
messages lost or cluster creation procedures that don't make sure
everybody is targeted by at least one incoming MEET message.
Also fix the logging of the IP switch avoiding the :-1 tail.
Also explicitly set version to 0, add a protocol version define, improve
comments in the gossip structure.
Note that the structure layout is the same after the change, we are just
making the padding explicit with an additional not used 16 bits field.
So this commit is still able to talk with the previous versions of
cluster nodes.
Valgrind checks that the buffers we transfer via syscalls are all
composed of bytes actually initialized. This is useful, it makes we able
to avoid leaking informations in non initialized parts fo messages
transferred to other hosts. This commit fixes one of such issues.
Can't be initialized by resetManualFailover() since it's actual state
the function uses, so we need to initialize it at startup time. Not
really a bug in practical terms, but showed up into valgrind and is not
technically correct anyway.
Adds configuration option 'supervised [no | upstart | systemd | auto]'
Also removed 'bzero' from the previous implementation because it's 2015.
(We could actually statically initialize those structs, but clang
throws an invalid warning when we try, so it looks bad even though it
isn't bad.)
Fixes#2264
Previously, Redis only wrote the pid file if
it was daemonizing, but many times it's useful to have
the pid written out even if you're in the foreground.
Some background for this is:
I usually run redis via daemontools. That entails running
redis-server on the foreground. Given that, I'd also want
redis-server to create a pidfile so other processes (e.g. nagios)
can run checks for that.
Closes#463
This commit introduces a new RDB data type called 'aux'. It is used in
order to insert inside an RDB file key-value pairs that may serve
different needs, without breaking backward compatibility when new
informations are embedded inside an RDB file. The contract between Redis
versions is to ignore unknown aux fields when encountered.
Aux fields can be used in order to:
1. Augment the RDB file with info like version of Redis that created the
RDB file, creation time, used memory while the RDB was created, and so
forth.
2. Add state about Redis inside the RDB file that we need to reload
later: replication offset, previos master run ID, in order to improve
failovers safety and allow partial resynchronization after a slave
restart.
3. Anything that we may want to add to RDB files without breaking the
ability of past versions of Redis to load the file.
The new opcode is an hint about the size of the dataset (keys and number
of expires) we are going to load for a given Redis database inside the
RDB file. Since hash tables are resized accordingly ASAP, useless
rehashing is avoided, speeding up load times significantly, in the order
of ~ 20% or more for larger data sets.
Related issue: #1719
Adds: ql_compressed (boolean, 1 if compression enabled for list, 0
otherwise)
Adds: ql_uncompressed_size (actual uncompressed size of all quicklistNodes)
Adds: ql_ziplist_max (quicklist max ziplist fill factor)
Compression ratio of the list is then ql_uncompressed_size / serializedlength
We report ql_uncompressed_size for all quicklists because serializedlength
is a _compressed_ representation anyway.
Sample output from a large list:
127.0.0.1:6379> llen abc
(integer) 38370061
127.0.0.1:6379> debug object abc
Value at:0x7ff97b51d140 refcount:1 encoding:quicklist serializedlength:19878335 lru:9718164 lru_seconds_idle:5 ql_nodes:21945 ql_avg_node:1748.46 ql_ziplist_max:-2 ql_compressed:0 ql_uncompressed_size:1643187761
(1.36s)
The 1.36s result time is because rdbSavedObjectLen() is serializing the
object, not because of any new stats reporting.
If we run DEBUG OBJECT on a compressed list, DEBUG OBJECT takes almost *zero*
time because rdbSavedObjectLen() reuses already-compressed ziplists:
127.0.0.1:6379> debug object abc
Value at:0x7fe5c5800040 refcount:1 encoding:quicklist serializedlength:19878335 lru:9718109 lru_seconds_idle:5 ql_nodes:21945 ql_avg_node:1748.46 ql_ziplist_max:-2 ql_compressed:1 ql_uncompressed_size:1643187761
This removes:
- list-max-ziplist-entries
- list-max-ziplist-value
This adds:
- list-max-ziplist-size
- list-compress-depth
Also updates config file with new sections and updates
tests to use quicklist settings instead of old list settings.
Let user set how many nodes to *not* compress.
We can specify a compression "depth" of how many nodes
to leave uncompressed on each end of the quicklist.
Depth 0 = disable compression.
Depth 1 = only leave head/tail uncompressed.
- (read as: "skip 1 node on each end of the list before compressing")
Depth 2 = leave head, head->next, tail->prev, tail uncompressed.
- ("skip 2 nodes on each end of the list before compressing")
Depth 3 = Depth 2 + head->next->next + tail->prev->prev
- ("skip 3 nodes...")
etc.
This also:
- updates RDB storage to use native quicklist compression (if node is
already compressed) instead of uncompressing, generating the RDB string,
then re-compressing the quicklist node.
- internalizes the "fill" parameter for the quicklist so we don't
need to pass it to _every_ function. Now it's just a property of
the list.
- allows a runtime-configurable compression option, so we can
expose a compresion parameter in the configuration file if people
want to trade slight request-per-second performance for up to 90%+
memory savings in some situations.
- updates the quicklist tests to do multiple passes: 200k+ tests now.
Added field 'ql_nodes' and 'ql_avg_per_node'.
ql_nodes is the number of quicklist nodes in the quicklist.
ql_avg_node is the average fill level in each quicklist node. (LLEN / QL_NODES)
Sample output:
127.0.0.1:6379> DEBUG object b
Value at:0x7fa42bf2fed0 refcount:1 encoding:quicklist serializedlength:18489 lru:8983768 lru_seconds_idle:3 ql_nodes:430 ql_avg_per_node:511.73
127.0.0.1:6379> llen b
(integer) 220044
Turns out it's a huge improvement during save/reload/migrate/restore
because, with compression enabled, we're compressing 4k or 8k
chunks of data consisting of multiple elements in one ziplist
instead of compressing series of smaller individual elements.
Use the existing memory space for an SDS to convert it to a regular
character buffer so we don't need to allocate duplicate space just
to extract a usable buffer for native operations.
Fill factor now has two options:
- negative (1-5) for size-based ziplist filling
- positive for length-based ziplist filling with implicit size cap.
Negative offsets define ziplist size limits of:
-1: 4k
-2: 8k
-3: 16k
-4: 32k
-5: 64k
Positive offsets now automatically limit their max size to 8k. Any
elements larger than 8k will be in individual nodes.
Positive ziplist fill factors will keep adding elements
to a ziplist until one of:
- ziplist has FILL number of elements
- or -
- ziplist grows above our ziplist max size (currently 8k)
When using positive fill factors, if you insert a large
element (over 8k), that element will automatically allocate
an individual quicklist node with one element and no other elements will be
in the same ziplist inside that quicklist node.
When using negative fill factors, elements up to the size
limit can be added to one quicklist node. If an element
is added larger than the max ziplist size, that element
will be allocated an individual ziplist in a new quicklist node.
Tests also updated to start testing at fill factor -5.
This started out as #2158 by sunheehnus, but I kept rewriting it
until I could understand things more easily and get a few more
correctness guarantees out of the readability flow.
The original commit created and returned a new ziplist with the contents of
both input ziplists, but I prefer to grow one of the input ziplists
and destroy the other one.
So, instead of malloc+copy as in #2158, the merge now reallocs one of
the existing ziplists and copies the other ziplist into the new space.
Also added merge test cases to ziplistTest()
This replaces individual ziplist vs. linkedlist representations
for Redis list operations.
Big thanks for all the reviews and feedback from everybody in
https://github.com/antirez/redis/pull/2143
The previous test wasn't returning the new ziplist, so the test
was invalid. Now the test works properly.
These problems were simultaenously discovered in #2154 and that
PR also had an additional fix we included here.
zipEntry was returning a struct, but that caused some
problems with tests under 32 bit builds.
The tests run better if we operate on structs allocated in the
caller without worrying about copying on return.
Previously, many files had individual main() functions for testing,
but each required being compiled with their own testing flags.
That gets difficult when you have 8 different flags you need
to set just to run all tests (plus, some test files required
other files to be compiled aaginst them, and it seems some didn't
build at all without including the rest of Redis).
Now all individual test main() funcions are renamed to a test
function for the file itself and one global REDIS_TEST define enables
testing across the entire codebase.
Tests can now be run with:
- `./redis-server test <test>`
e.g. ./redis-server test ziplist
If REDIS_TEST is not defined, then no tests get included and no
tests are included in the final redis-server binary.
1. Server unxtime may remain not updated while loading AOF, so ETA is
not updated correctly.
2. Number of processed byte was not initialized.
3. Possible division by zero condition (likely cause of issue #1932).
1. memory leak in t_set.c has been fixed
2. end-of-line spaces has been removed (from all over the place)
3. for loops have been ordered up to match existing Redis style (less weird)
4. comments format has been fixed (added * in the beggining of every comment line)
setTypeRandomElements() now returns unsigned long, and also uses unsigned long for anything related to count of members.
spopWithCountCommand() now uses unsigned long elements_returned instead of int, for values returned from setTypeRandomElements()
If we woke up to accept a connection, but we can't
accept it, inform the user of the error going on
with their networking.
(The previous message was the same for success or error!)
spopCommand() now runs spopWithCountCommand() in case the <count> param is found.
Added intsetRandomMembers() to Intset: Copies N random members from the set into inputted 'values' array. Uses either the Knuth or Floyd sample algos depending on ratio count/size.
Added setTypeRandomElements() to SET type: Returns a number of random elements from a non empty set. This is a version of setTypeRandomElement() that is modified in order to return multiple entries, using dictGetRandomKeys() and intsetRandomMembers().
Added tests for SPOP with <count>: unit/type/set, unit/scripting, integration/aof
--
Cleaned up code a bit to match with required Redis coding style
Otherwise there are security risks, especially when providing Redis as a
service, the user may "sniff" for admin commands renamed to an
unguessable string via rename-command in redis.conf.
Previously the string was created empty then re-sized
to fit the offset, but sds resize causes the sds to
over-allocate by at least 1 MB (which is a lot when
you are operating at bit-level access).
This also improves the speed of initial sets by 2% to 6%
based on quick testing.
Patch logic provided by @oranagra
Fixes#1918
Improvements:
- Return empty string if asking for non-existing section (INFO foo)
- Fix potential memory leak (caused by sdsempty() then returned if >2 args)
- Clean up argument parsing
- Allow "all" as valid section (same as "default" or zero args currently)
- Move strcasecmp to end of evaluation chain in conditionals
Also, since we're C99, I moved some variable declarations to be closer
to where they are actually used (saves us from needing to free an empty info
if detect argument errors up front).
Closes#1915Closes#1966
There is no standard cross-platform way of obtaining
system memory info, but I found a useful function
convering all common platforms. I removed support
for uncommon Redis platforms (windows, AIX) and left
others intact.
For more info, see:
http://nadeausoftware.com/articles/2012/09/c_c_tip_how_get_physical_memory_size_system
The system memory info is cached on startup, but some systems
may be able to change the amount of memory visible to Redis
at runtime if Redis is deployed in a VM or container.
Also see #1820
Slaves key expire is orchestrated by the master. Sometimes the master
will send the synthesized DEL to expire keys on the slave with a non
trivial delay (when the key is not accessed, only the incremental expiry
algorithm will expire it in background).
During that time, a key is logically expired, but slaves still return
the key if you GET (or whatever) it. This is a bad behavior.
However we can't simply trust the slave view of the key, since we need
the master to be able to send write commands to update the slave data
set, and DELs should only happen when the key is expired in the master
in order to ensure consistency.
However 99.99% of the issues with this behavior is when a client which
is not a master sends a read only command. In this case we are safe and
can consider the key as non existing.
This commit does a few changes in order to make this sane:
1. lookupKeyRead() is modified in order to return NULL if the above
conditions are met.
2. Calls to lookupKeyRead() in commands actually writing to the data set
are repliaced with calls to lookupKeyWrite().
There are redundand checks, so for example, if in "2" something was
overlooked, we should be still safe, since anyway, when the master
writes the behavior is to don't care about what expireIfneeded()
returns.
This commit is related to #1768, #1770, #2131.
I guess the initial goal of the initialization was to suppress GCC
warning, but if we have to initialize, we can do it with the base-case
value instead of NULL which is never retained.
in the case (all chars of the string s found in 'cset' ),
line[573] will no more do the same thing line[572] did.
this will be more faster especially in the case that the string s is very long and all chars of string s found in 'cset'
Track bandwidth used by clients and replication (but diskless
replication is not tracked since the actual transfer happens in the
child process).
This includes a refactoring that makes tracking new instantaneous
metrics simpler.
PFCOUNT is technically speaking a write command, since the cached value
of the HLL is exposed in the data structure (design error, mea culpa), and
can be modified by PFCOUNT.
However if we flag PFCOUNT as "w", read only slaves can't execute the
command, which is a problem since there are environments where slaves
are used to scale PFCOUNT reads.
Nor it is possible to just prevent PFCOUNT to modify the data structure
in slaves, since without the cache we lose too much efficiency.
So while this commit allows slaves to create a temporary inconsistency
(the strings representing the HLLs in the master and slave can be
different in certain moments) it is actually harmless.
In the long run this should be probably fixed by turning the HLL into a
more opaque representation, for example by storing the cached value in
the part of the string which is not exposed (this should be possible
with SDS strings).
bulk_data field size was not removed from the count. It is not possible
to declare it simply as 'char bulk_data[]' since the structure is nested
into another structure.
Because of (not so) recent Redis changes, now the LRU internally
reported unit is milliseconds, not seconds, but the DEBUG OBJECT output
was still claiming seconds while providing milliseconds.
However OBJECT IDLETIME was working as expected, which is the correct
API to use.
zmalloc(0) cauesd to actually trigger a non-zero allocation since with
standard libc malloc we have our own zmalloc header for memory tracking,
but at the same time the returned pointer is at the end of the block and
not in the middle. This triggers a false positive when testing with
valgrind.
When the inline protocol args count is 0, we now avoid reallocating
c->argv, preventing the issue to happen.
Issue: #2157
As the SET command is parsed, it remembers which options are already set
and if a duplicate option is found, raises an error because it is
essentially an invalid syntax.
It still allows mutually exclusive options like EX and PX because taking
an option over another (precedence) is not essentially a syntactic
error.
Sentinel queries the INFO from every master and from every replica of
every master.
We can cache the INFO results in Sentinel so Sentinel can be a single
place to quickly get all INFO output for an entire Sentinel monitoring
group.
This commit gives us SENTINEL INFO-CACHE in two forms:
- SENTINEL INFO-CACHE — returns all masters and all replicas
- SENTINEL INFO-CACHE master0 master1 ... masterN — vararg specify masters
Results are returned as a multibulk reply with two top-level entries
for each master. The first entry for each master is the name of the master.
The second entry is a nested multibulk reply with the contents of INFO,
first for the master, then an additional entry for each of the
replicas.
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.
Both upstart and systemd provide a way for daemons to
be supervised, as well as a mechanism for them to
signal their readyness status.
This patch provides compatibility with this functionality while
not interfering with other methods.
With this, it will be possible to use `expect stop` with upstart
and `Type=notify` with systemd.
A more detailed explanation of the mechanism can be found here:
http://spootnik.org/entries/2014/11/09_pid-tracking-in-modern-init-systems.html
if redis works in cluster-mode and redis-cli was run with argv, reconnect if needs.
example:
./redis-cli set foo bar
if return is MOVED redis-cli just do nothing.
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).
People mostly use SORT against lists, but our prior
behavior was pretending lists were an unordered bag
requiring a forced-sort when no sort was requested.
We can just use the native list ordering to ensure
consistency across replicaion and scripting calls.
Closes#2079Closes#545 (again)
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.
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.
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.
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.
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.
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.
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.
A few people have written custom C commands because bit
manipulation isn't exposed through Lua. Let's give
them Mike Pall's bitop.
This adds bitop 1.0.2 (2012-05-08) from http://bitop.luajit.org/
bitop is imported as "bit" into the global namespace.
New Lua commands: bit.tobit, bit.tohex, bit.bnot, bit.band, bit.bor, bit.bxor,
bit.lshift, bit.rshift, bit.arshift, bit.rol, bit.ror, bit.bswap
Verification of working (the asserts would abort on error, so (nil) is correct):
127.0.0.1:6379> eval "assert(bit.tobit(1) == 1); assert(bit.band(1) == 1); assert(bit.bxor(1,2) == 3); assert(bit.bor(1,2,4,8,16,32,64,128) == 255)" 0
(nil)
127.0.0.1:6379> eval 'assert(0x7fffffff == 2147483647, "broken hex literals"); assert(0xffffffff == -1 or 0xffffffff == 2^32-1, "broken hex literals"); assert(tostring(-1) == "-1", "broken tostring()"); assert(tostring(0xffffffff) == "-1" or tostring(0xffffffff) == "4294967295", "broken tostring()")' 0
(nil)
Tests also integrated into the scripting tests and can be run with:
./runtest --single unit/scripting
Tests are excerpted from `bittest.lua` included in the bitop distribution.
With the exception of nodes sending MEET packets: we have to trust them
since they can send us MEET packets only when the cluster is initially
created or because sysadmin manual action.
In the cluster evaluation function we are supposed to set the cluster
state as "fail" if we are among a minority, however the code was not
detecting to be into a minority partition if exactly half the masters
were reachable, which is a minority.
We need to remember what is the saving strategy of the current RDB child
process, since the configuration may be modified at runtime via CONFIG
SET and still we'll need to understand, when the child exists, what to
do and for what goal the process was initiated: to create an RDB file
on disk or to write stuff directly to slave's sockets.
However we don't try to do this if the integer is already inside a range
representable with a shared integer.
The performance gain appears to be around ~15% in micro benchmarks,
however in the long run this also helps to improve locality, so should
have more, hard to measure, benefits.
Some language in the comment was difficult
to understand, so this commit: clarifies wording, removes
unnecessary words, and relocates some dependent clauses
closer to what they actually describe.
I also tried to break up longer chains of thought
(if X, then Y, and Q, and also F, so obviously M)
into more manageable chunks for ease of understanding.
- Remove trailing newlines from redis.conf
- Fix comment misspelling
- Clarifies zipEncodeLength usage and a C API mention (#1243, #1242)
- Fix cluster typos (inspired by @papanikge #1507)
- Fix rewite -> rewrite in a few places (inspired by #682)
Closes#1243, #1242, #1507
The old DEBUG POPULATE form for automatic creation of test keys is:
DEBUG POPULATE <count>
Now an additional form is available:
DEBUG POPULATE <count> <prefix>
When prefix is not specified, it defaults to "key", so the keys are
named incrementally from key:0 to key:<count-1>. Otherwise the specified
prefix is used instead of "key".
The command is useful in order to populate different Redis instances
with key names guaranteed to don't collide. There are other debugging
uses, for example it is possible to add additional N keys using a count
of N and a random prefix at every call.
Following the CLIENT LIST output format, we prefix the unix socket
address with a "/" so that it is different than an IPv4/6 address.
This makes parsing simpler.
Related to #2010.
This fixes a potential bug that was never observed in practice since
what happens is that the asynchronous connect returns ok (to fail later,
calling the handler) every time, so a ping is queued, and sent_ping
happens to always be populated.
Howver technically connect(2) with a non blocking socket may return an
error synchronously, so before this fix the code was not correct.
It is not clear if files open in append only mode will automatically fix
their offset after a truncate(2) operation. This commit makes sure that
we reposition the AOF file descriptor offset at the end of the file
after a truncated AOF is loaded and trimmed to the last valid command.
Recently we introduced the ability to load truncated AOFs, but
unfortuantely the support was broken since the server, after loading the
truncated AOF, continues appending to the file that is corrupted at the
end. The problem is fixed only in the next AOF rewrite.
This commit fixes the issue by truncating the AOF to the last valid
opcode, and aborting if it is not possible to truncate the file
correctly.
The code to check the number of voters was never updated to follow the new
Sentinel specification, so the number of voters was computed using only
the set of Sentinels that provided a vote.
This means that there is a changing majority on partitions, even if
usually the issue is not triggered because of the configured quorum
check (what was broken was the other implicit check that requires anyway
half of the known sentinels to agree in order to start a failover).
Because of the new ability to start with a truncated AOF, we need
to correctly release all the memory on EOF error. Otherwise there is a
small leak, that is not really a problem, but causes a false positive in
the tests that detect memory leaks.
