MySQL master discovery methods, part 6: other methods

This is the sixth in a series of posts reviewing methods for MySQL master discovery: the means by which an application connects to the master of a replication tree. Moreover, the means by which, upon master failover, it identifies and connects to the newly promoted master.

These posts are not concerned with the manner by which the replication failure detection and recovery take place. I will share orchestrator specific configuration/advice, and point out where cross DC orchestrator/raft setup plays part in discovery itself, but for the most part any recovery tool such as MHA, replication-manager, severalnines or other, is applicable.

Hard coded configuration deployment

You may use your source/config repo as master service discovery method of sorts.

The master’s identity would be hard coded into your, say, git repo, to be updated and deployed to production upon failover.

This method is simple and I’ve seen it being used by companies, in production. Noteworthy:

Continue reading » “MySQL master discovery methods, part 6: other methods”

MySQL master discovery methods, part 5: Service discovery & Proxy

This is the fifth in a series of posts reviewing methods for MySQL master discovery: the means by which an application connects to the master of a replication tree. Moreover, the means by which, upon master failover, it identifies and connects to the newly promoted master.

These posts are not concerned with the manner by which the replication failure detection and recovery take place. I will share orchestrator specific configuration/advice, and point out where cross DC orchestrator/raft setup plays part in discovery itself, but for the most part any recovery tool such as MHA, replication-manager, severalnines or other, is applicable.

We discuss asynchronous (or semi-synchronous) replication, a classic single-master-multiple-replicas setup. A later post will briefly discuss synchronous replication (Galera/XtraDB Cluster/InnoDB Cluster).

Master discovery via Service discovery and Proxy

Part 4 presented with an anti-pattern setup, where a proxy would infer the identify of the master by drawing conclusions from backend server checks. This led to split brains and undesired scenarios. The problem was the loss of context.

We re-introduce a service discovery component (illustrated in part 3), such that:

  • The app does not own the discovery, and
  • The proxy behaves in an expected and consistent way.

In a failover/service discovery/proxy setup, there is clear ownership of duties:

  • The failover tool own the failover itself and the master identity change notification.
  • The service discovery component is the source of truth as for the identity of the master of a cluster.
  • The proxy routes traffic but does not make routing decisions.
  • The app only ever connects to a single target, but should allow for a brief outage while failover takes place.

Depending on the technologies used, we can further achieve:

  • Hard cut for connections to old, demoted master M.
  • Black/hold off for incoming queries for the duration of failover.

We explain the setup using the following assumptions and scenarios:

  • All clients connect to master via cluster1-writer.example.net, which resolves to a proxy box.
  • We fail over from master M to promoted replica R.

Continue reading » “MySQL master discovery methods, part 5: Service discovery & Proxy”

MySQL master discovery methods, part 4: Proxy heuristics

Note: the method described here is an anti pattern

This is the fourth in a series of posts reviewing methods for MySQL master discovery: the means by which an application connects to the master of a replication tree. Moreover, the means by which, upon master failover, it identifies and connects to the newly promoted master.

These posts are not concerned with the manner by which the replication failure detection and recovery take place. I will share orchestrator specific configuration/advice, and point out where cross DC orchestrator/raft setup plays part in discovery itself, but for the most part any recovery tool such as MHA, replication-manager, severalnines or other, is applicable.

We discuss asynchronous (or semi-synchronous) replication, a classic single-master-multiple-replicas setup. A later post will briefly discuss synchronous replication (Galera/XtraDB Cluster/InnoDB Cluster).

Master discovery via Proxy Heuristics

In Proxy Heuristics all clients connect to the master through a proxy. The proxy observes the backend MySQL servers and determines who the master is.

This setup is simple and easy, but is an anti pattern. I recommend against using this method, as explained shortly.

Clients are all configured to connect to, say, cluster1-writer.proxy.example.net:3306. The proxy will intercept incoming requests either based on hostname or by port. It is aware of all/some MySQL backend servers in that cluster, and will route traffic to the master M.

A simple heuristic that I’ve seen in use is: pick the server that has read_only=0, a very simple check.

Let’s take a look at how this works and what can go wrong.

Continue reading » “MySQL master discovery methods, part 4: Proxy heuristics”

MySQL master discovery methods, part 3: app & service discovery

This is the third in a series of posts reviewing methods for MySQL master discovery: the means by which an application connects to the master of a replication tree. Moreover, the means by which, upon master failover, it identifies and connects to the newly promoted master.

These posts are not concerned with the manner by which the replication failure detection and recovery take place. I will share orchestrator specific configuration/advice, and point out where cross DC orchestrator/raft setup plays part in discovery itself, but for the most part any recovery tool such as MHA, replication-manager, severalnines or other, is applicable.

We discuss asynchronous (or semi-synchronous) replication, a classic single-master-multiple-replicas setup. A later post will briefly discuss synchronous replication (Galera/XtraDB Cluster/InnoDB Cluster).

App & service discovery

Part 1 and part 2 presented solutions where the app remained ingorant of master’s identity. This part takes a complete opposite direction and gives the app ownership on master access.

We introduce a service discovery component. Commonly known are Consul, ZooKeeper, etcd, highly available stores offering key/value (K/V) access, leader election or full blown service discovery & health.

We satisfy ourselves with K/V functionality. A key would be mysql/master/cluster1 and a value would be the master’s hostname/port.

It is the app’s responsibility at all times to fetch the identity of the master of a given cluster by querying the service discovery component, thereby opening connections to the indicated master.

The service discovery component is expected to be up at all times and to contain the identity of the master for any given cluster.

Continue reading » “MySQL master discovery methods, part 3: app & service discovery”

MySQL master discovery methods, part 2: VIP & DNS

This is the second in a series of posts reviewing methods for MySQL master discovery: the means by which an application connects to the master of a replication tree. Moreover, the means by which, upon master failover, it identifies and connects to the newly promoted master.

These posts are not concerned with the manner by which the replication failure detection and recovery take place. I will share orchestrator specific configuration/advice, and point out where cross DC orchestrator/raft setup plays part in discovery itself, but for the most part any recovery tool such as MHA, replication-manager, severalnines or other, is applicable.

We discuss asynchronous (or semi-synchronous) replication, a classic single-master-multiple-replicas setup. A later post will briefly discuss synchronous replication (Galera/XtraDB Cluster/InnoDB Cluster).

Master discovery via VIP

In part 1 we saw that one the main drawbacks of DNS discovery is the time it takes for the apps to connect to the promoted master. This is the result of both DNS deployment time as well as client’s TTL.

A quicker method is offered: use of VIPs (Virtual IPs). As before, apps would connect to cluster1-writer.example.net, cluster2-writer.example.net, etc. However, these would resolve to specific VIPs.

Say cluster1-writer.example.net resolves to 10.10.0.1. We let this address float between servers. Each server has its own IP (say 10.20.0.XXX) but could also potentially claim the VIP 10.10.0.1.

VIPs can be assigned by switches and I will not dwell into the internals, because I’m not a network expert. However, the following holds:

  • Acquiring a VIP is a very quick operation.
  • Acquiring a VIP must take place on the acquiring host.
  • A host may be unable to acquire a VIP should another host holds the same VIP.
  • A VIP can only be assigned within a bounded space: hosts connected to the same switch; hosts in the same Data Center or availability zone.

Continue reading » “MySQL master discovery methods, part 2: VIP & DNS”

MySQL master discovery methods, part 1: DNS

This is the first in a series of posts reviewing methods for MySQL master discovery: the means by which an application connects to the master of a replication tree. Moreover, the means by which, upon master failover, it identifies and connects to the newly promoted master.

These posts are not concerned with the manner by which the replication failure detection and recovery take place. I will share orchestrator specific configuration/advice, and point out where cross DC orchestrator/raft setup plays part in discovery itself, but for the most part any recovery tool such as MHA, replication-manager, severalnines or other, is applicable.

We discuss asynchronous (or semi-synchronous) replication, a classic single-master-multiple-replicas setup. A later post will briefly discuss synchronous replication (Galera/XtraDB Cluster/InnoDB Cluster).

Master discovery via DNS

In DNS master discovery applications connect to the master via a name that gets resolved to the master’s box. By way of example, apps would target the masters of different clusters by connecting to cluster1-writer.example.net, cluster2-writer.example.net, etc. It is up for the DNS to resolve those names to IPs.

Continue reading » “MySQL master discovery methods, part 1: DNS”

Using dbdeployer in CI tests

I was very pleased when Giuseppe Maxia (aka datacharmer) unveiled dbdeployer in his talk at pre-FOSDEM MySQL day. The announcement came just at the right time. I wish to briefly describe how we use dbdeployer (work in progress).

The case for gh-ost

A user opened an issue on gh-ost, and the user was using MySQL 5.5. gh-ost is being tested on 5.7 where the problem does not reproduce. A discussion with Gillian Gunson raised the concern of not testing on all versions. Can we run gh-ost tests for all MySQL/Percona/MariaDB versions? Should we? How easy would it be?

gh-ost tests

gh-ost has three different test types:

  • Unit tests: these are plain golang logic tests which are very easy and quick to run.
  • Integration tests: the topic of this post, see following. Today these do not run as part of an automated CI testing.
  • System tests: putting our production tables to the test, continuously migrating our production data on dedicated replicas, verifying checksums are identical and data is intact, read more.

Unit tests are already running as part of automated CI (every PR is subjected to those tests). Systems tests are clearly tied to our production servers. What’s the deal with the integration tests? Continue reading » “Using dbdeployer in CI tests”

orchestrator 3.0.6: faster crash detection & recoveries, auto Pseudo-GTID, semi-sync and more

orchestrator 3.0.6 is released and includes some exciting improvements and features. It quickly follows up on 3.0.5 released recently, and this post gives a breakdown of some notable changes:

Faster failure detection

Recall that orchestrator uses a holistic approach for failure detection: it reads state not only from the failed server (e.g. master) but also from its replicas. orchestrator now detects failure faster than before:

  • A detection cycle has been eliminated, leading to quicker resolution of a failure. On our setup, where we poll servers every 5sec, failure detection time dropped from 7-10sec to 3-5sec, keeping reliability. The reduction in time does not lead to increased false positives.
    Side note: you may see increased not-quite-failure analysis such as “I can’t see the master” (UnreachableMaster).
  • Better handling of network scenarios where packets are dropped. Instead of hanging till TCP timeout, orchestrator now observes server discovery asynchronously. We have specialized failover tests that simulate dropped packets. The change reduces detection time by some 5sec.

Faster master recoveries

Promoting a new master is a complex task which attempts to promote the best replica out of the pool of replicas. It’s not always the most up-to-date replica. The choice varies depending on replica configuration, version, and state.

With recent changes, orchestrator is able to to recognize, early on, that the replica it would like to promote as master is ideal. Assuming that is the case, orchestrator is able to immediate promote it (i.e. run hooks, set read_only=0 etc.), and run the rest of the failover logic, i.e. the rewiring of replicas under the newly promoted master, asynchronously.

This allows the promoted server to take writes sooner, even while its replicas are not yet connected. It also means external hooks are executed sooner.

Between faster detection and faster recoveries, we’re looking at some 10sec reduction in overall recovery time: from moment of crash to moment where a new master accepts writes. We stand now at < 20sec in almost all cases, and < 15s in optimal cases. Those times are measured on our failover tests.

We are working on reducing failover time unrelated to orchestrator and hope to update soon.

Automated Pseudo-GTID

As reminder, Pseudo-GTID is an alternative to GTID, without the kind of commitment you make with GTID. It provides similar “point your replica under any other server” behavior GTID allows. Continue reading » “orchestrator 3.0.6: faster crash detection & recoveries, auto Pseudo-GTID, semi-sync and more”

Implementing non re-entrant functions in Golang

A non re-entrant function is a function that could only be executing once at any point in time, regardless of how many times it is being invoked and by how many goroutines.

This post illustrates blocking non re-entrant functions and yielding non re-entrant functions implementations in golang.

A use case

A service is polling for some conditions, monitoring some statuses once per second. We want each status to be checked independently of others without blocking. An implementation might look like:

func main() {
    tick := time.Tick(time.Second)
    go func() {
        for range tick {
            go CheckSomeStatus()
            go CheckAnotherStatus()
        }
    }()
}

We choose to run each status check in its own goroutine so that CheckAnotherStatus() doesn’t wait upon CheckSomeStatus() to complete.

Each of these checks typically take a very short amount of time, and much less than a second. What happens, though, if CheckAnotherStatus() itself takes more than one second to run? Perhaps there’s an unexpected network or disk latency affecting the execution time of the check.

Does it make sense for the function to be executed twice at the same time? If not, we want it to be non re-entrant. Continue reading » “Implementing non re-entrant functions in Golang”

orchestrator 3.0.3: auto provisioning raft nodes, native Consul support and more

orchestrator 3.0.3 is released! There’s been a lot going on since 3.0.2:

orchestrator/raft: auto-provisioning nodes via lightweight snaphsots

In an orchestrator/raft setup, we have n hosts forming a raft cluster. In a 3-node setup, for example, one node can go down, and still the remaining two will form a consensus, keeping the service operational. What happens when the failed node returns?

With 3.0.3 the failed node can go down for as long as it wants. Once it comes back, it attempts to join the raft cluster. A node keeps its own snapshots and its raft log outside the relational backend DB. If it has recent-enough data, it just needs to catch up with raft replication log, which is acquires from one of the active nodes.

If its data is very stale, it will request a snapshot from an active node, which it will import, and will just resume from that point.

If its data is gone, that’s not a problem. It gets a snapshot from an active node, improts it, and keeps running from that point.

If it’s a newly provisioned box, that’s not a problem. It gets a snapshot from an active node, … etc.

  • SQLite backed setups can just bootstrap new nodes. No need to dump+load or import any data.
    • Side effect: you may actually use :memory:, where SQLite does not persist any data to disk. Remember that the raft snapshots and replication log will cover you. The cheat is that the raft replication log itself is managed and persisted by an independent SQLite database.
  • MySQL backed setups will still need to make sure orchestrator has the privileges to deploy itself.

More info in the docs.

This plays very nicely into the hands of kubernetes, which is on orchestrator‘s roadmap.

Key Value, native Consul support (Zk TODO)

orchestrator now supports Key-Value stores built-in, and Consul in particular.

At this time the purpose of orchestrator KV is to support master discovery. orchestrator will write the identity of the master of each cluster to KV store. The user will use that information to apply changes to their infrastructure.

For example, the user will rely on Consul KV entries, written by orchestrator, to generate proxy config files via consul-template, such that traffic is directed via the proxy onto the correct master.

orchestrator supports:

  • Manually writing identity of cluster’s master to KV store
    • e.g. `orchestrator-client -c submit-masters-to-kv-stores -alias mycluster`
  • Automatically updating master’s identify upon failover

Key-value pairs are in the form of <cluster-alias>&lt;master&gt;. For example:

  • Key is `main_cluster`
  • Value is my-db-0123.my.company.com:3306

Web UI improvements

Using the web UI, you can now: Continue reading » “orchestrator 3.0.3: auto provisioning raft nodes, native Consul support and more”