The MANIFESTO is not Valkey's manifesto and it doesn't even mention open
source. Let's write another one later, or some other document about our
project principles.
The other two files are one-line files with no relevant info. They're
polluting the file listing at root level. It's the first thing you see
when you start exploring the repo for the first time.
Signed-off-by: Viktor Söderqvist <viktor.soderqvist@est.tech>
Migrate the contents in TLS.md into TLS sections including building,
running and detail supports. TODO list in the TLS.md is almost done
except the implementation of benchmark support is still not the best
approach which should migrate to hiredis async mode.
Closes#888
---------
Signed-off-by: I Hsin Cheng <richard120310@gmail.com>
Signed-off-by: Viktor Söderqvist <viktor.soderqvist@est.tech>
Co-authored-by: Viktor Söderqvist <viktor.soderqvist@est.tech>
In #885, we only add a shutdown path, there is another path
is that the server might got hang by slowlog. This PR added
the pause path coverage to cover it.
Signed-off-by: Binbin <binloveplay1314@qq.com>
## Description
When I explore the cycles distributions for `lrange` test (
`valkey-benchmark -p 9001 -t lrange -d 100 -r 1000000 -n 1000000 -c 50
--threads 4`). I found the `prepareClientToWrite` and
`clientHasPendingReplies` could be reduced to single call outside
instead of called in a loop, ideally we can gain 3% performance. The
corresponding `LRANG_100`, `LRANG_300`, `LRANGE_500`, `LRANGE_600` have
~2% - 3% performance boost, the benchmark test prove it helps.
This patch try to move the `prepareClientToWrite` and its child
`clientHasPendingReplies` out of the loop to reduce the function
overhead.
---------
Signed-off-by: Lipeng Zhu <lipeng.zhu@intel.com>
The test might be fast enough and then there is no change in the role
causing the test to fail. Adding a wait to avoid the timing issue:
```
*** [err]: valkey-cli make source node ignores NOREPLICAS error when doing the last CLUSTER SETSLOT
Expected '{127.0.0.1 23154 267}' to be equal to '' (context: type eval line 24 cmd {assert_equal [lindex [R 3 role] 2] {}} proc ::test)
```
Signed-off-by: Binbin <binloveplay1314@qq.com>
sdsAllocSize returns the correct size without consulting the
allocator. Which is much faster than consulting the allocator.
The only exception is SDS_TYPE_5, for which it has to
consult the allocator.
This PR also sets alloc field correctly for embedded string objects.
It assumes that no allocator would allocate a buffer larger
than `259 + sizeof(robj)` for embedded string. We use embedded strings
for strings up to 44 bytes. If this assumption is wrong, the whole
function would require a rewrite. In general case sds type adjustment
might be needed. Such logic should go to sds.c.
---------
Signed-off-by: Vadym Khoptynets <vadymkh@amazon.com>
Currently, the `dual-channel-replication` feature flag is immutable if
`enable-protected-configs` is enabled, which is the default behavior.
This PR proposes to make the `dual-channel-replication` flag mutable,
allowing it to be changed dynamically without restarting the cluster.
**Motivation:**
The ability to change the `dual-channel-replication` flag dynamically is
essential for testing and validating the feature on real clusters
running in production environments. By making the flag mutable, we can
enable or disable the feature without disrupting the cluster's
operations, facilitating easier testing and experimentation.
Additionally, this change would provide more flexibility for users to
enable or disable the feature based on their specific requirements or
operational needs without requiring a cluster restart.
---------
Signed-off-by: naglera <anagler123@gmail.com>
1. Remove redundant connIncrRefs/connDecrRefs
In socket.c, the reference counter is incremented before calling
callHandler, but the same reference counter is also incremented inside
callHandler before calling the actual callback.
static inline int callHandler(connection *conn, ConnectionCallbackFunc handler) {
connIncrRefs(conn);
if (handler) handler(conn);
connDecrRefs(conn);
...
}
This commit removes the redundant incr/decr calls in socket.c
2. Correct return value of connRead for TLS when peer closed
According to comments in connection.h, connRead returns 0 when the peer
has closed the connection. This patch corrects the return value for TLS
connections. (Without this patch, it returns -1 which means error.)
There is an observable difference in what is logged in the verbose
level: "Client closed connection" vs "Reading from client: (null)".
---------
Signed-off-by: Viktor Söderqvist <viktor.soderqvist@est.tech>
This PR utilizes the IO threads to execute commands in batches, allowing
us to prefetch the dictionary data in advance.
After making the IO threads asynchronous and offloading more work to
them in the first 2 PRs, the `lookupKey` function becomes a main
bottle-neck and it takes about 50% of the main-thread time (Tested with
SET command). This is because the Valkey dictionary is a straightforward
but inefficient chained hash implementation. While traversing the hash
linked lists, every access to either a dictEntry structure, pointer to
key, or a value object requires, with high probability, an expensive
external memory access.
### Memory Access Amortization
Memory Access Amortization (MAA) is a technique designed to optimize the
performance of dynamic data structures by reducing the impact of memory
access latency. It is applicable when multiple operations need to be
executed concurrently. The principle behind it is that for certain
dynamic data structures, executing operations in a batch is more
efficient than executing each one separately.
Rather than executing operations sequentially, this approach interleaves
the execution of all operations. This is done in such a way that
whenever a memory access is required during an operation, the program
prefetches the necessary memory and transitions to another operation.
This ensures that when one operation is blocked awaiting memory access,
other memory accesses are executed in parallel, thereby reducing the
average access latency.
We applied this method in the development of `dictPrefetch`, which takes
as parameters a vector of keys and dictionaries. It ensures that all
memory addresses required to execute dictionary operations for these
keys are loaded into the L1-L3 caches when executing commands.
Essentially, `dictPrefetch` is an interleaved execution of dictFind for
all the keys.
**Implementation details**
When the main thread iterates over the `clients-pending-io-read`, for
clients with ready-to-execute commands (i.e., clients for which the IO
thread has parsed the commands), a batch of up to 16 commands is
created. Initially, the command's argv, which were allocated by the IO
thread, is prefetched to the main thread's L1 cache. Subsequently, all
the dict entries and values required for the commands are prefetched
from the dictionary before the command execution. Only then will the
commands be executed.
---------
Signed-off-by: Uri Yagelnik <uriy@amazon.com>
This example was for script replication which we have
completely removed in 7.0, so this example is outdated now.
Signed-off-by: Binbin <binloveplay1314@qq.com>
Although KEYS is a dangerous command and we recommend people
to avoid using it, some people who are not familiar with it
still using it, and even use KEYS with no pattern at all.
Once KEYS is using with no pattern, we can convert it to an
exact match to avoid iterating over all data.
Signed-off-by: Binbin <binloveplay1314@qq.com>
According to the Python document[1], any invalid escape sequences in
string literals now generate a DeprecationWarning (SyntaxWarning as of
3.12) and in the future this will become a SyntaxError.
This Change uses Python’s raw string notation for regular expression
patterns to avoid it.
[1]: https://docs.python.org/3.10/library/re.html
Signed-off-by: haoqixu <hq.xu0o0@gmail.com>
When failover deny to vote, sometimes due to network or
some blocking operations, the time of FAILOVER_AUTH_REQUEST
packet arrival is very uncertain. Since there is no epoch
information in these logs, it is hard to associate the log
with other logs.
Signed-off-by: Binbin <binloveplay1314@qq.com>
At the VERBOSE/DEBUG log level, which is output once every 5 seconds,
added to show the "Total" message of all clients and to show memory
usage (used_memory) with used_memory_human.
Also, it seems clearer to show "total" number of keys and the number of
volatile in entire keys.
---------
Signed-off-by: NAM UK KIM <namuk2004@naver.com>
Add new optional, immutable string config called `unixsocketgroup`.
Change the group of the unix socket to `unixsocketgroup` after `bind()`
if specified.
Adds tests to validate the behavior.
Fixes#873.
Signed-off-by: Ayush Sharma <mrayushs933@gmail.com>
Today if we attached the "run-extra-tests" tag it adds at least 20
minutes because the dump-fuzzer test runs with full accuracy. This
fuzzer is useful, but probably only really needed for the daily, so
removing it from the PRs. We still run the fuzzers, just not for as
long.
Signed-off-by: Madelyn Olson <madelyneolson@gmail.com>
This fixes#899. In that issue, the primary is cluster-allow-replica-migration no
and its replica is cluster-allow-replica-migration yes.
And during the slot migration:
1. Primary calling blockClientForReplicaAck, waiting its replica.
2. Its replica reconfiguring itself as a replica of other shards due to
replica migration and disconnect from the old primary.
3. The old primary never got the chance to receive the ack, so it got a
timeout and got a NOREPLICAS error.
In this case, the replicas might automatically migrate to another primary,
resulting in the client being unblocked with the NOREPLICAS error. In this
case, since the configuration will eventually propagate itself, we can safely
ignore this error on the source node.
Signed-off-by: Binbin <binloveplay1314@qq.com>
In CLUSTER SETSLOT propagation logic, if the replicas are down, the
client will get block during command processing and then unblock
with `NOREPLICAS Not enough good replicas to write`.
The reason is that all replicas are down (or some are down), but
myself->num_replicas is including all replicas, so the client will
get block and always get timeout.
We should only wait for those online replicas, otherwise the waiting
propagation will always timeout since there are not enough replicas.
The admin can easily check if there are replicas that are down for an
extended period of time. If they decide to move forward anyways, we
should not block it. If a replica failed right before the replication and
was not included in the replication, it would also unlikely win the election.
Signed-off-by: Binbin <binloveplay1314@qq.com>
Co-authored-by: Ping Xie <pingxie@google.com>
There is a test that assumes that the backlog will get overrun, but
because of the recent changes to the default it no longer fails. It
seems like it is a bit flakey now though, so resetting the value in the
test back to 1mb. (This relates to the CoB of 1100k. So it should
consistently work with a 1mb limit).
Signed-off-by: Madelyn Olson <madelyneolson@gmail.com>
Now, when clients run the unsubscribe, sunsubscribe and punsubscribe
commands in the non-subscribed mode, it returns 0.
Indeed this is a bug, we should not allow client run these kind of
commands here.
Thus, this PR fixes this bug, but it is a break change for existing
clients
---------
Signed-off-by: hwware <wen.hui.ware@gmail.com>
In #764, we add a --io-threads mode in test, but forgot
to handle runtest-cluster, they are different framework.
Currently runtest-cluster does not support tags, and we
don't have plan to support it. And currently cluster tests
does not have any io-threads tests, so this PR just align
--io-threads option with #764.
Signed-off-by: Binbin <binloveplay1314@qq.com>
If n is already myself primary, there is no need to re-establish the
replication connection.
In the past we allow a replica node to reconnect with its primary via
this CLUSTER REPLICATE command, it will use psync. But since #885, we
will assume that a full sync is needed in this case, so if we don't do
this, the replica will always use full sync.
Signed-off-by: Binbin <binloveplay1314@qq.com>
Co-authored-by: Ping Xie <pingxie@google.com>
Fix feedback loop in key eviction with tracking clients when using I/O
threads.
Current issue:
Evicting keys while tracking clients or key space-notification exist
creates a feedback loop when using I/O threads:
While evicting keys we send tracking async writes to I/O threads,
preventing immediate release of tracking clients' COB memory
consumption.
Before the I/O thread finishes its write, we recheck used_memory, which
now includes the tracking clients' COB and thus continue to evict more
keys.
**Fix:**
We will skip the test for now while IO threads are active. We may
consider avoiding sending writes in `processPendingWrites` to I/O
threads for tracking clients when we are out of memory.
---------
Signed-off-by: Uri Yagelnik <uriy@amazon.com>
Signed-off-by: Madelyn Olson <madelyneolson@gmail.com>
Co-authored-by: Madelyn Olson <madelyneolson@gmail.com>
Update README.md since the project is no longer under construction, and
can reference the main website.
---------
Signed-off-by: Harkrishn Patro <harkrisp@amazon.com>
The repl-backlog-size 1mb is too small in most cases, now network
transmission and bandwidth performance have improved rapidly in more
than ten years.
The bigger the replication backlog, the longer the replica can endure
the disconnect and later be able to perform a partial resynchronization.
Part of #653.
---------
Signed-off-by: Binbin <binloveplay1314@qq.com>
Just add one more test for command "sentinel IS-PRIMARY-DOWN-BY-ADDR" to
make the reply-schemas-validator
run successfully.
Note: test result here
https://github.com/hwware/valkey/actions/runs/10457516111
Signed-off-by: hwware <wen.hui.ware@gmail.com>
If the client side crashes by any issue or exits normally, the kernel
will try to disconnect RDMA QPs. Then the kernel of server side receives
CM packets, valkey-server handles CM disconnected event and close
connection.
However, there is a lack of keepalive mechanism from RDMA transport
layer. Once the kernel of client side crashes, the server side will not
be notified. To avoid this issue, valkey server sents Keepaliv command
periodically to detect any dead QPs.
An example of mlx-cx5:
```
# RDMA: CQ handle error status: transport retry counter exceeded[0xc], opcode : 0x0
# RDMA: CQ handle error status: transport retry counter exceeded[0xc], opcode : 0x0
# RDMA: CQ handle error status: Work Request Flushed Error[0x5], opcode : 0x0
# RDMA: CQ handle error status: Work Request Flushed Error[0x5], opcode : 0x0
# RDMA: CQ handle error status: Work Request Flushed Error[0x5], opcode : 0x0
# RDMA: CQ handle error status: Work Request Flushed Error[0x5], opcode : 0x0
```
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Our current replica can initiate a failover without restriction when
it detects that the primary node is offline. This is generally not a
problem. However, consider the following scenarios:
1. In slot migration, a primary loses its last slot and then becomes
a replica. When it is fully synchronized with the new primary, the new
primary downs.
2. In CLUSTER REPLICATE command, a replica becomes a replica of another
primary. When it is fully synchronized with the new primary, the new
primary downs.
In the above scenario, case 1 may cause the empty primary to be elected
as the new primary, resulting in primary data loss. Case 2 may cause the
non-empty replica to be elected as the new primary, resulting in data
loss and confusion.
The reason is that we have cached primary logic, which is used for psync.
In the above scenario, when clusterSetPrimary is called, myself will cache
server.primary in server.cached_primary for psync. In replicationGetReplicaOffset,
we get server.cached_primary->reploff for offset, gossip it and rank it,
which causes the replica to use the old historical offset to initiate
failover, and it get a good rank, initiates election first, and then is
elected as the new primary.
The main problem here is that when the replica has not completed full
sync, it may get the historical offset in replicationGetReplicaOffset.
The fix is to clear cached_primary in these places where full sync is
obviously needed, and let the replica use offset == 0 to participate
in the election. In this way, this unhealthy replica has a worse rank
and is not easy to be elected.
Of course, it is possible that it will be elected with offset == 0.
In the future, we may need to prohibit the replica with offset == 0
from having the right to initiate elections.
Another point worth mentioning, in above cases:
1. In the ROLE command, the replica status will be handshake, and the
offset will be -1.
2. Before this PR, in the CLUSTER SHARD command, the replica status will
be online, and the offset will be the old cached value (which is wrong).
3. After this PR, in the CLUSTER SHARD, the replica status will be loading,
and the offset will be 0.
Signed-off-by: Binbin <binloveplay1314@qq.com>
In these places we should use RDB_EOF_MARK_SIZE, but we mixed
it with CONFIG_RUN_ID_SIZE. This is not an issue since they are
all 40, just a cleanup.
Signed-off-by: Binbin <binloveplay1314@qq.com>
Saves some memory (one pointer) in the client struct.
Since a client cannot be blocked multiple times, we can assume
it will be held in only one extra utility list, so it is ok to maintain
a union of these listNode references.
Signed-off-by: Ran Shidlansik <ranshid@amazon.com>
Signed-off-by: ranshid <88133677+ranshid@users.noreply.github.com>
Signed-off-by: Binbin <binloveplay1314@qq.com>
Co-authored-by: Viktor Söderqvist <viktor.soderqvist@est.tech>
Co-authored-by: Binbin <binloveplay1314@qq.com>
Currently, if the client enters a blocked state, it will be
added to the server.clients_waiting_acks list. When the client
is unblocked, that is, when unblockClient is called, we will
need to linearly traverse server.clients_waiting_acks to delete
the client, and this search is O(N).
When WAIT (or WAITAOF) is used extensively in some cases, this
O(N) search may be time-consuming. We can remember the list node
and store it in the blockingState struct and it can avoid the
linear search in unblockClientWaitingReplicas.
Signed-off-by: Binbin <binloveplay1314@qq.com>
Add 4 new commands for Sentinel (reference
https://github.com/valkey-io/valkey/issues/36)
Sentinel GET-PRIMARY-ADDR-BY-NAME
Sentinel PRIMARY
Sentinel PRIMARIES
Sentinel IS-PRIMARY-DOWN-BY-ADDR
and deprecate 4 old commands:
Sentinel GET-MASTER-ADDR-BY-NAME
Sentinel MASTER
Sentinel MASTERS
Sentinel IS-MASTER-DOWN-BY-ADDR
and all sentinel tests pass here
https://github.com/hwware/valkey/actions/runs/9962102363/job/27525124583
Note:
1. runtest-sentinel pass all test cases
2. I finished a sentinel rolling upgrade test: 1 primary 2 replicas 3
sentinel
there are 4 steps in this test scenario:
step 1: all 3 sentinel nodes run old sentinel, shutdown primary, and
then new primary can be voted successfully.
step 2: replace sentinel 1 with new sentinel bin file, and then shutdown
primary, and then another new primary can be voted successfully
step 3: replace sentinel 2 with new sentinel bin file, and then shutdown
primary, and then another new primary can be voted successfully
step 4: replace sentinel 3 with new sentinel bin file, and then shutdown
primary, and then another new primary can be voted successfully
We can see, even mixed version sentinel running, whole system still
works.
---------
Signed-off-by: hwware <wen.hui.ware@gmail.com>
For `debug object` command, we use `val->lru` but ignore the `lfu` mode.
So in `lfu` mode, `debug object` would return meaningless `lru` descriptions.
Added two new fields lfu_freq and lfu_access_time_minutes.
Signed-off-by: jiangyujie.jyj <yjjiang1996@163.com>
Co-authored-by: Binbin <binloveplay1314@qq.com>
Adding FAST option to DEBUG OBJECT command.
The light version only shows the light weight infomation,
which mostly O(1). The pre-existing version that show more
stats such as serializedlength sometimes is time consuming.
This should allow looking into debug stats (the key expired
but not deleted), even on huge object, on which we're afraid
to run the command for fear of causing a server freeze.
Somehow like 3ca451c46fed894bf49e7561fa0282d2583f1c06.
Signed-off-by: Binbin <binloveplay1314@qq.com>
In #786, we did skip it in the daily, but not for the others.
When running ./runtest on MacOS, we will get the failure.
```
couldn't open socket: host is unreachable (nodename nor servname provided, or not known)
```
The reason is that TCL 8.5 doesn't support ipv6, so we skip tests
tagged with ipv6. This also revert #786.
Signed-off-by: Binbin <binloveplay1314@qq.com>
I've tried to test a dual channel replication but forgot to add +sync
for my replication user. As a result replica entered silent cycle like
this:
```
* Connecting to PRIMARY 127.0.0.1:6379
* PRIMARY <-> REPLICA sync started
* Non blocking connect for SYNC fired the event.
* Primary replied to PING, replication can continue...
* Trying a partial resynchronization (request ...)
* PSYNC is not possible, initialize RDB channel.
* Aborting dual channel sync
```
And primary got endless cycle like this:
```
* Replica 127.0.0.1:6380 asks for synchronization
* Partial resynchronization not accepted: Replication ID mismatch (Replica asked for '...', my replication IDs are '...' and '...')
* Replica 127.0.0.1:6380 is capable of dual channel synchronization, and partial sync isn't possible. Full sync will continue with dedicated RDB channel.
```
There was no way to understand that replication user is missing +sync
acl on notice log level. With this one-line change we get a warning
message in our replica log.
---------
Signed-off-by: secwall <secwall@yandex-team.ru>
Update references of copyright being assigned to Salvatore when it was
transferred to Redis Ltd. as per
https://github.com/valkey-io/valkey/issues/544.
---------
Signed-off-by: Pieter Cailliau <pieter@redis.com>
A pointer to dtype is stored in the dict forever.
dtype is stack-allocated while the dict created is global.
The dict (and the pointer to dtype in it) will live past the lifetime of
dtype.
clusterManagerLinkDictType is a global object that has the same values
as dtype.
Signed-off-by: Salvatore Mesoraca <salvatore.mesoraca@aiven.io>
Fix#821
During the `FAILOVER` process, when conditions are met (such as when the
force time is reached or the primary and replica offsets are
consistent), the primary actively becomes the replica and transitions to
the `FAILOVER_IN_PROGRESS` state. After the primary becomes the replica,
and after handshaking and other operations, it will eventually send the
`PSYNC FAILOVER` command to the replica, after which the replica will
become the primary. This means that the upgrade of the replica to the
primary is an asynchronous operation, which implies that during the
`FAILOVER_IN_PROGRESS` state, there may be a period of time where both
nodes are replicas. In this scenario, if a `-REDIRECT` is returned, the
request will be redirected to the replica and then redirected back,
causing back and forth redirection. To avoid this situation, during the
`FAILOVER_IN_PROGRESS state`, we temporarily suspend the clients that
need to be redirected until the replica truly becomes the primary, and
then resume the execution.
---------
Signed-off-by: zhaozhao.zz <zhaozhao.zz@alibaba-inc.com>
We are updating this variable in the main thread, and the
child threads can printing the logs at the same time. This
generating a warning in SANITIZER=thread:
```
WARNING: ThreadSanitizer: data race (pid=74208)
Read of size 4 at 0x000102875c10 by thread T3:
#0 serverLogRaw <null>:52173615 (valkey-server:x86_64+0x10003c556)
#1 _serverLog <null>:52173615 (valkey-server:x86_64+0x10003ca89)
#2 bioProcessBackgroundJobs <null>:52173615 (valkey-server:x86_64+0x1001402c9)
Previous write of size 4 at 0x000102875c10 by main thread (mutexes: write M0):
#0 afterSleep <null>:52173615 (valkey-server:x86_64+0x10004989b)
#1 aeProcessEvents <null>:52173615 (valkey-server:x86_64+0x100031e52)
#2 main <null>:52173615 (valkey-server:x86_64+0x100064a3c)
#3 start <null>:52173615 (dyld:x86_64+0xfffffffffff5c365)
#4 start <null>:52173615 (dyld:x86_64+0xfffffffffff5c365)
```
The refresh of daylight_active is not real time, we update
it in aftersleep, so we don't need a strong synchronization,
so using memory_order_relaxed. But also noted we are doing
load/store operations only for daylight_active, which is an
aligned 32-bit integer, so using memory_order_relaxed will
not provide more consistency than what we have today.
So this is just a cleanup that to clear the warning.
Signed-off-by: Binbin <binloveplay1314@qq.com>
Test dual-channel-replication primary gets cob overrun during replica
rdb load` fails during the Valgrind run. This is due to the load
handlers disconnecting before the tests complete, resulting in a low
primary COB. Increasing the handlers' timeout should resolve this issue.
Failure:
https://github.com/valkey-io/valkey/actions/runs/10361286333/job/28681321393
Server logs reveals that the load handler clients were disconnected
before the test started
Also the two previus test took about 20 seconds which is the handler
timeout.
---------
Signed-off-by: naglera <anagler123@gmail.com>
Signed-off-by: Madelyn Olson <madelyneolson@gmail.com>
Co-authored-by: Madelyn Olson <madelyneolson@gmail.com>
The clusterManagerCompareKeysValues is introduced in
143bfa1e6e65cf8be1eaad0b8169e2d95ca62f9a, which calls
DEBUG DIGEST-VALUE to check whether the value of the
source node and the target node are consistent.
However, the DEBUG DIGEST-VALUE command is not supported
in older version, such as version 4, and the node will return
unknown subcommand. Or the DEBUG command can be disabled
in version 7, and the node will return DEBUG not allowed.
In these cases, we need to output friendly message to
allow users to proceed to the next step, instead of just
outputing `Value check failed!`.
Unknown subcommand example:
```
*** Target key exists
*** Checking key values on both nodes...
Node 127.0.0.1:30001 replied with error:
ERR unknown subcommand or wrong number of arguments for 'DIGEST-VALUE'. Try DEBUG HELP.
Node 127.0.0.1:30003 replied with error:
ERR unknown subcommand or wrong number of arguments for 'DIGEST-VALUE'. Try DEBUG HELP.
*** Value check failed!
DEBUG DIGEST-VALUE command is not supported.
You can relaunch the command with --cluster-replace option to force key overriding.
```
DEBUG not allowed example:
```
*** Target key exists
*** Checking key values on both nodes...
Node 127.0.0.1:30001 replied with error:
ERR DEBUG command not allowed. If the enable-debug-command option is ...
Node 127.0.0.1:30003 replied with error:
ERR DEBUG command not allowed. If the enable-debug-command option is ...
*** Value check failed!
DEBUG command is not allowed.
You can turn on the enable-debug-command option.
Or you can relaunch the command with --cluster-replace option to force key overriding.
```
Signed-off-by: Binbin <binloveplay1314@qq.com>
In the past implementation of `ZUNION` and `ZUNIONSTORE` commands, we
first create a temporary dict called `accumulator`. After adding all
member-score mappings to `accumulator`, we still need to convert
`accumulator` back to the final dict `dstzset->dict`. However, we can
directly use `dstzset->dict` to avoid the additional copy operation.
This PR removes the `accumulator` dict and directly uses` dstzset->dict
`to store the member-score mappings.
- **Test**
First, I added 300 unique elements to two sorted sets called
'zunion_test1' and 'zunion_test2'. Then, I tested `zunion` and
`zunionstore` on these two sorted sets. The test results shown below
indicate that the performance of both zunion and zunionstore improved
about 31%.
### ZUNION
#### unstable
```
./valkey-benchmark -P 10 -n 100000 zunion 2 zunion_test1 zunion_test2
Summary:
throughput summary: 2713.41 requests per second
latency summary (msec):
avg min p50 p95 p99 max
146.252 3.464 153.343 182.015 184.959 192.895
```
#### This PR
```
./valkey-benchmark -P 10 -n 100000 zunion 2 zunion_test1 zunion_test2
Summary:
throughput summary: 3543.84 requests per second
latency summary (msec):
avg min p50 p95 p99 max
108.259 2.984 114.239 141.695 145.151 160.255
```
### ZUNIONSTORE
#### unstable
```
./valkey-benchmark -P 10 -n 100000 zunionstore out 2 zunion_test1 zunion_test2
Summary:
throughput summary: 3168.07 requests per second
latency summary (msec):
avg min p50 p95 p99 max
157.511 3.368 183.167 189.311 193.535 231.679
```
#### This PR
```
./valkey-benchmark -P 10 -n 100000 zunionstore out 2 zunion_test1 zunion_test2
Summary:
throughput summary: 4144.73 requests per second
latency summary (msec):
avg min p50 p95 p99 max
120.374 2.648 141.823 149.119 153.855 183.167
```
---------
Signed-off-by: RayCao <zisong.cw@alibaba-inc.com>
Signed-off-by: zisong.cw <zisong.cw@alibaba-inc.com>
A configuration option with zero impact on server operation but is
printed out on server crash and can be accessed by gdb for debugging. It
can be used by the user/operator to store any free-form string. This
string will persist as long as the server is running and will be
accessible in the following ways:
And printed in crash reports:
```
------ CONFIG DEBUG OUTPUT ------
lazyfree-lazy-eviction no
...
io-threads-do-reads yes
debug-context "test2"
proto-max-bulk-len 512mb
```
---------
Signed-off-by: Eran Liberty <eranl@amazon.com>
Co-authored-by: Eran Liberty <eranl@amazon.com>
- Fix TLS bug where connection were shutdown by primary's main process
while the child process was still writing- causing main process to be
blocked.
- TLS connection fix -file descriptors are set to blocking mode in the
main thread, followed by a blocking write. This sets the file
descriptors to non-blocking if TLS is used (see `connTLSSyncWrite()`)
(@xbasel).
- Improve the reliability of dual-channel tests. Modify the pause
mechanism to verify process status directly, rather than relying on log.
- Ensure that `server.repl_offset` and `server.replid` are updated
correctly when dual channel synchronization completes successfully.
Thist led to failures in replication tests that validate replication IDs
or compare replication offsets.
---------
Signed-off-by: naglera <anagler123@gmail.com>
Signed-off-by: naglera <58042354+naglera@users.noreply.github.com>
Signed-off-by: xbasel <103044017+xbasel@users.noreply.github.com>
Signed-off-by: Madelyn Olson <madelyneolson@gmail.com>
Signed-off-by: Binbin <binloveplay1314@qq.com>
Co-authored-by: ranshid <88133677+ranshid@users.noreply.github.com>
Co-authored-by: xbasel <103044017+xbasel@users.noreply.github.com>
Co-authored-by: Madelyn Olson <madelyneolson@gmail.com>
Co-authored-by: Binbin <binloveplay1314@qq.com>
