// Open source database support consolidation
One database support contract for PostgreSQL, MySQL, MongoDB and Redis
Database support from OSSeva is one contract, one escalation path and one renewal date for the open-source databases you run yourself: PostgreSQL, MySQL, MariaDB, MongoDB, Redis and Valkey, plus the Elasticsearch and messaging clusters beside them, including versions the community no longer patches. Critical CVEs (CVSS 9.0 or higher) are fixed within 48 hours and High within 7 days, Assure adds 24/7 reactive support, and Operate adds proactive monitoring and a 15-minute P1 response. Enter your work email for a quote, or read the coverage table below.
- Engines
- PostgreSQL, MySQL, MariaDB, MongoDB, Redis and Valkey, Elasticsearch, plus Kafka, RabbitMQ and ZooKeeper
- Versions
- Any covered line, end-of-life versions included, through signed patched builds
- Escalation
- One path across every engine, with a named senior engineer on Operate
- Patch window
- Critical CVEs (CVSS 9.0 or higher) within 48 hours, High within 7 days
- Support hours
- Assure includes 24/7 reactive support; Operate adds the 15-minute P1
- Pricing model
- Per cluster, across every engine on the contract. OSSeva does not host databases
End-of-life dates checked on endoflife.date and each project's lifecycle policy on 20 October 2026; next review 18 January 2027.
Trusted globally by enterprises





Engines and versions covered by one database support contract
One contract covers every engine below, on your own servers, VMs, cloud account or Kubernetes. Lines whose community support has ended stay covered through signed, patched builds, so consolidating does not wait for an upgrade. End dates checked 20 October 2026 on endoflife.date.
| Engine | Lines OSSeva supports | Already past community end of life |
|---|---|---|
| PostgreSQL | 11 to 18 | 11 (9 Nov 2023), 12 (21 Nov 2024), 13 (13 Nov 2025). 14 ends 12 Nov 2026. |
| MySQL | 5.7, 8.0, 8.4 LTS, 9.7 LTS | 5.7 (31 Oct 2023), 8.0 (30 Apr 2026) |
| MariaDB | 10.4 to 10.6, 10.11, 11.4, 11.8, 12.3 | 10.4 (18 Jun 2024), 10.5 (24 Jun 2025), 10.6 (6 Jul 2026) |
| MongoDB | 4.2 to 6.0, self-managed, and current lines | 4.2 (30 Apr 2023), 4.4 (29 Feb 2024), 5.0 (31 Oct 2024), 6.0 (31 Jul 2025) |
| Redis and Valkey | Redis 6.2 and 7.0, Valkey 7.2, 8.x and 9.x | Redis 7.0 (29 Jul 2024). Redis 6.2 ends 1 Apr 2027. |
| Elasticsearch | 7.10.2, 7.17 and 8.x | 7.x (15 Jan 2026) |
The same contract also covers the messaging and coordination layer around the databases: Kafka, RabbitMQ, ZooKeeper, etcd, ClickHouse and Apache Pulsar. OSSeva does not support Microsoft SQL Server or Oracle Database, and it does not host databases. Lifecycle dates: endoflife.date and each project's lifecycle policy.
Every supported version, by technology →End-of-life tracker →
Response times by severity, one table for every engine
The targets do not change with the engine. A PostgreSQL outage and a Redis outage are both P1 and get the same response, which is the practical difference from holding one contract per product.
| Priority | Definition | Response | Resolution target |
|---|---|---|---|
| P1 | Production down or data loss risk | 15 minutes | 4 hours |
| P2 | Significant degradation, no immediate outage | 1 hour | 8 hours |
| P3 | Non-critical issue, workaround available | 4 hours | 2 business days |
| P4 | Question, advisory, enhancement request | 1 business day | Scheduled sprint |
- Patch: signed patched builds for every covered line. Critical CVEs (CVSS 9.0 or higher) within 48 hours and High within 7 days.
- Assure: everything in Patch, plus 24/7 reactive support. Raise an issue at any hour and an OSSeva engineer works it with you. Assure does not monitor or run your clusters.
- Operate: everything in Patch and Assure, plus proactive 24/7 monitoring, the response and resolution targets above, and a named senior engineer on your account.
The response and resolution targets are from the OSSeva Operate service description.
One escalation path for every database and the services around it
When a checkout slows down, the cause can sit in the database, the cache or the broker. With one vendor per product, each checks its own piece and your team runs the investigation. With one contract, the investigation has one owner.
- You raise the issue once, whichever engine it looks like. There is no decision about which vendor to call first.
- On Operate, the named senior engineer on your account owns it end to end and brings in the engineer for each engine involved. On Assure, an OSSeva engineer works it with you at any hour.
- If the cause crosses engines, one OSSeva engineer holds the investigation instead of three vendors each ruling out their own product.
- Serious issues go to senior OSSeva engineers through the on-call escalation that Operate includes.
- Every incident ends with a post-mortem and a prevention plan, written once for the whole estate.
Escalation to senior engineers, the named engineer and post-mortems are Operate inclusions. Assure provides 24/7 reactive support without them.
What changes when you consolidate database support
The databases, binaries and data stay where they are. What changes is the paperwork and who answers the phone.
| One contract per product | One OSSeva contract | |
|---|---|---|
| Contracts and renewals | A contract and a renewal date for each product, each with its own procurement cycle. | One agreement and one renewal date. Adding an engine later goes onto the same contract. |
| Price model | Per core, per node or per GB, so growth in one place changes several bills. | Per cluster, across every engine on the contract. Adding capacity inside a cluster does not change the support bill. |
| Versions past end of life | Often outside the contract, on self-support, with nobody shipping fixes. | Covered through signed patched builds, in the table above. |
| Incidents across engines | Each vendor checks its own product. | One escalation path and one owner for the investigation. |
| Audit evidence | SBOMs and patch records from several sources in several formats. | SBOMs and CVE remediation records from one source, on Assure. |
| Getting there | Wait for the first renewal, or pay twice while contracts overlap. | OSSeva picks up each technology as its current contract ends, so there is no gap and no long overlap. |
Consolidation is not the right move for everyone. If you run one engine and want it operated for you, managed PostgreSQL on your own servers is a closer fit, and if you depend on features that exist only in a vendor's own distribution, staying with that vendor may be better.
Switching vendors without migrating →Managed PostgreSQL on your own servers →Compare your spend with the cost calculator →
Worked example: three engines, three ended lines, one inventory
An example estate, not a customer: PostgreSQL 13, MongoDB 6.0 and Redis 7.0, each past community end of life. Since those dates the site's CVE directory lists 18 CVEs, all scored 7.0 or higher, that either name the line or, for PostgreSQL 13, were fixed only on newer lines, counted 20 October 2026. Three vendor contracts that stop at upstream support would each leave their line on self-support. One contract lists all three in the same version map and covers them with patched builds.
| Line | Community end of life | CVEs in the directory since | Highest CVSS | How counted |
|---|---|---|---|---|
| PostgreSQL 13 | 13 Nov 2025 | 8 | 8.8 | Fixed on 14 to 18 in the 2026 minor releases. PostgreSQL does not assess 13, so no release exists for it. |
| MongoDB 6.0 | 31 Jul 2025 | 3 | 8.8 | The advisory's affected range names 6.0. |
| Redis 7.0 | 29 Jul 2024 | 7 | 9.9 | The affected range reads "before" a later fix and so reaches 7.0. |
PostgreSQL 13: CVEs published after 13 Nov 2025
Show all 8 entries
| CVE | Published | CVSS | Affected range (NVD) | Fixed upstream in |
|---|---|---|---|---|
| CVE-2026-14680 | 2026-08-13 | 8.8 | Before 18.6, 17.11, 16.15, 15.19 and 14.24 (supported versions only; 13 and older not assessed) | 18.6; 17.11; 16.15; 15.19; 14.24 |
| CVE-2026-16239 | 2026-08-13 | 8.8 | Before 18.6, 17.11, 16.15, 15.19 and 14.24 (supported versions only; 13 and older not assessed) | 18.6; 17.11; 16.15; 15.19; 14.24 |
| CVE-2026-15741 | 2026-08-13 | 8.8 | Before 18.6, 17.11, 16.15, 15.19 and 14.24 (supported versions only; 13 and older not assessed) | 18.6; 17.11; 16.15; 15.19; 14.24 |
| CVE-2026-19385 | 2026-08-13 | 8.8 | Before 18.6, 17.11, 16.15, 15.19 and 14.24 (supported versions only; 13 and older not assessed) | 18.6; 17.11; 16.15; 15.19; 14.24 |
| CVE-2026-6473 | 2026-05-14 | 8.8 | Before 18.4, 17.10, 16.14, 15.18 and 14.23 (supported versions only; 13 and older not assessed) | 18.4; 17.10; 16.14; 15.18; 14.23 |
| CVE-2026-6477 | 2026-05-14 | 8.8 | Before 18.4, 17.10, 16.14, 15.18 and 14.23 (supported versions only; 13 and older not assessed) | 18.4; 17.10; 16.14; 15.18; 14.23 |
| CVE-2026-2006 | 2026-02-12 | 8.8 | Before 18.2, 17.8, 16.12, 15.16 and 14.21 (supported versions only; 13 and older not assessed) | 18.2; 17.8; 16.12; 15.16; 14.21 |
| CVE-2026-6479 | 2026-05-14 | 7.5 | Before 18.4, 17.10, 16.14, 15.18 and 14.23 (supported versions only; 13 and older not assessed) | 18.4; 17.10; 16.14; 15.18; 14.23 |
MongoDB 6.0: CVEs published after 31 Jul 2025
Show all 3 entries
| CVE | Published | CVSS | Affected range (NVD) | Fixed upstream in |
|---|---|---|---|---|
| CVE-2026-11933 | 2026-06-12 | 8.8 | 8.3.3 and earlier; 8.2.10 and earlier; 8.0.25 and earlier; 7.0.36 and earlier; 6.0.28 and earlier; 5.0.33 and earlier; 4.4.30 and earlier | 8.3.4; 8.2.12; 8.0.26; 7.0.37; 6.0.29; 5.0.34; 4.4.31 |
| CVE-2026-8053 | 2026-05-13 | 8.8 | 8.3 before 8.3.2; 8.2 before 8.2.9; 8.0 before 8.0.23; 7.0 before 7.0.34; 6.0 before 6.0.28; 5.0 before 5.0.33 | 8.3.2; 8.2.9; 8.0.23; 7.0.34; 6.0.28; 5.0.33 |
| CVE-2025-14847 | 2025-12-19 | 7.5 | 8.2.0 to 8.2.2; 8.0.0 to 8.0.16; 7.0 before 7.0.28; 6.0.0 to 6.0.26; 5.0.0 to 5.0.31; 4.4.0 to 4.4.29; all 4.2, 4.0 and 3.6 versions | 8.2.3; 8.0.17; 7.0.28; 6.0.27; 5.0.32; 4.4.30 |
Redis 7.0: CVEs published after 29 Jul 2024
Show all 7 entries
| CVE | Published | CVSS | Affected range (NVD) | Fixed upstream in |
|---|---|---|---|---|
| CVE-2024-31449 | 2024-10-07 | 8.8 | 2.8.18 and later, before 6.2.16, 7.2.6 and 7.4.1 | 6.2.16, 7.2.6, 7.4.1 |
| CVE-2024-46981 | 2025-01-06 | 9.8 | All versions with Lua scripting before 6.2.17, 7.2.7 and 7.4.2 | 6.2.17, 7.2.7, 7.4.2 |
| CVE-2025-21605 | 2025-04-23 | 7.5 | 2.6 and later, before 6.2.18, 7.2.8 and 7.4.3 | 6.2.18, 7.2.8, 7.4.3 |
| CVE-2025-32023 | 2025-07-07 | 7.8 | 2.8 and later, before 6.2.19, 7.2.10, 7.4.5 and 8.0.3 | 6.2.19, 7.2.10, 7.4.5, 8.0.3 |
| CVE-2025-48367 | 2025-07-07 | 7.5 | All versions before 6.2.19, 7.2.10, 7.4.5 and 8.0.3 | 6.2.19, 7.2.10, 7.4.5, 8.0.3 |
| CVE-2025-46817 | 2025-10-03 | 8.8 | All versions with Lua scripting before 6.2.20, 7.2.11, 7.4.6, 8.0.4 and 8.2.2 | 6.2.20, 7.2.11, 7.4.6, 8.0.4, 8.2.2 |
| CVE-2025-49844 | 2025-10-03 | 9.9 | <6.2.20; >=7.0 <7.2.11; >=7.4.0 <7.4.6; >=8.0.0 <8.0.4 | See advisory |
The directory lists CVEs with a CVSS score of 7.0 or higher that are matched at NVD to the product, so it is a subset of every advisory. A listed CVE is a fact about the project, not a finding about your servers. Scope for a real estate depends on the number of clusters, the versions and where they run.
The CVE directory →PostgreSQL security vulnerabilities after end of life →
What contract sprawl costs you
Each contract made sense when it was signed. Together they make every incident and every renewal harder.
Incidents cross vendor lines
A slow checkout can involve the database, the cache and the broker at once. With three vendors, each one checks its own piece, and your team ends up running the investigation.
Renewals never stop
Separate contracts mean separate renewal dates, price models and procurement cycles through the year. Some bill per core, some per node, some per GB, so growth in one place changes several bills.
Old versions fall outside every contract
Many vendors support only versions that upstream still maintains. The clusters you have not upgraded yet, often the most important ones, end up on self-support with nobody shipping fixes.
What one OSSeva contract delivers
Inventory and version map
Every cluster, its version, where it runs and which contract covers it today, with the end-of-life date for each version on one page.
Transition contract by contract
OSSeva picks up each technology as its current contract ends, so you never pay twice for long and never have a gap.
Patched builds, including ended lines
Signed builds for the versions you run, including lines upstream no longer patches, such as PostgreSQL 11 to 14, Redis 6.2 and 7.0, RabbitMQ 3.x, MongoDB 4.2 to 6.0, Elasticsearch 7.10.2 and 7.17, and ZooKeeper 3.4, 3.6 and 3.7 (3.5 as custom support scoped per engagement).
Upgrade planning and audit evidence
Upgrade and migration planning per technology, plus SBOMs and CVE remediation records your auditors can check.
One escalation path, around the clock
Assure includes 24/7 reactive support. Operate adds proactive monitoring across the whole stack, a named senior engineer and a 15-minute P1 response, with Critical CVEs (CVSS 9.0 or higher) patched within 48 hours and High within 7 days.
Runs where you run
Bare metal, VMs, any Kubernetes distribution and any cloud account. There is no platform to adopt first.
Choosing a database provider: one contract or one per engine
| Option | What you get | Trade-off |
|---|---|---|
| Keep a contract per technology | Deep specialists for each product | Several renewals, several price models, and incidents that cross vendor lines. |
| A single-database vendor | Strong depth on one engine, often with its own enhanced distribution | Caches, brokers and streaming still need other contracts, and coverage usually tracks upstream support dates. |
| Your cloud database provider | Hosted engines and support in the same bill | Only for workloads on that provider's services, on its version calendar. The provider also keeps the superuser account. |
| A database management company | Operations done for you, often on the company's platform | Moving databases onto someone else's platform is a project. OSSeva Operate runs them on your infrastructure and does not host. |
| OSSeva | One contract across the stack, including end-of-life versions, on your infrastructure | You run the infrastructure, or add OSSeva Operate for 24/7 operations. |
Coverage from the OSSeva technology pages for PostgreSQL, MySQL, MariaDB, MongoDB, Redis, Elasticsearch, RabbitMQ, Kafka and ZooKeeper, and the Patch, Assure and Operate service pages, checked 20 October 2026. End-of-life dates from endoflife.date.
Who offers the best database software support?
It depends on which engines you run and on which versions. For one engine on a current version, the vendor behind it, or your cloud provider for a hosted service, is often the simplest choice. For several open-source engines, and especially for versions the community has stopped patching, look for one contract that covers all of them, with published response times and patch windows. OSSeva's are in the tables above: Critical CVEs within 48 hours, High within 7 days, 24/7 reactive support on Assure and a 15-minute P1 on Operate.
Can one support contract cover PostgreSQL, MongoDB and Redis?
Yes. One OSSeva contract covers PostgreSQL, MongoDB and Redis together with MySQL, MariaDB, Valkey, Elasticsearch and the messaging layer, at the same patch windows and response times, with one escalation path and one renewal date. It is priced per cluster across every engine on the contract.
What is open source database support?
A commercial agreement that gives you fixes, security patches and expert help for open-source databases you run yourself, so you are not relying on community forums and upstream release dates alone. OSSeva covers the databases and the caches, brokers and streaming platforms around them under one contract.
What is a database provider, and how is it different from database support?
A database provider supplies the database itself, usually as a hosted service from a cloud company or as software from the vendor behind it. Database support is the fixes, patches and expert help for a database you run. OSSeva is not a database provider: it does not host or sell databases, and supports the open-source ones you run on your own servers, VMs, cloud account or Kubernetes.
Is OSSeva a database management company?
Partly. With OSSeva Operate, our engineers monitor and run your databases around the clock, which is what database management companies do. The difference is where: OSSeva works on your own infrastructure and does not host databases, so you keep the servers, the data and full access.
Which technologies does one OSSeva contract cover?
PostgreSQL, MySQL, MariaDB, Redis, Valkey, MongoDB, Elasticsearch with OpenSearch as a migration target, Apache Kafka, RabbitMQ, ZooKeeper, etcd, ClickHouse, Apache Pulsar, ActiveMQ and the runtime layer underneath, such as Erlang/OTP. OSSeva does not support Microsoft SQL Server or Oracle Database. The technologies page lists every version.
Do we have to upgrade before we consolidate?
No. OSSeva ships patched, signed builds for end-of-life versions, so you can move support first and upgrade on your own schedule.
What happens to our current contracts?
Nothing changes until each one ends. OSSeva picks up each technology as its current contract ends, so you do not pay twice for long and there is no gap. Bring your renewal dates to the discovery call and we map the order.
Who supports PostgreSQL?
The PostgreSQL Global Development Group maintains it and fixes each major version for five years. Commercial support comes from cloud providers for their managed services and from support companies, OSSeva among them, for PostgreSQL you run yourself.
How is it priced?
Per cluster, not per core, per GB or per node, so adding capacity does not change the support bill. Enter your work email for a quote, or compare your current spend with the database support cost calculator.
What is third-party database support?
Support from a company other than the one that sells or maintains the database. For open-source software that is normal: the code is public, and OSSeva builds, patches and supports it for you.
Who is this for?
Teams running three or more open-source data or messaging technologies on their own infrastructure who want fewer contracts, one renewal date and one place to call when something breaks. If you run a single engine, the page for that engine is a closer fit.
Get a quote
One contract, one renewal date, one escalation path.
Enter your work email with your inventory and renewal dates, and we will map a transition and a quote.
Prefer to talk or read more first?