Solace 24/7 Support & Incident Management
Your messaging team runs Solace, we assist
Best for: Teams operating Solace brokers who want an expert reachable for spool, failover or delivery problems.
Solace support, 24/7
AceMQ supports Solace 24/7 on software brokers, appliances and event mesh estates, beside your Solace plan or on its own. Most calls start with one of three problems: a message spool filling because one queue has no consumer, an HA pair that did not fail over cleanly, or a slow subscriber backing up a whole message VPN.
24/7 across Americas, EMEA and APAC, with response targets from 15 minutes, by plan, and a written root cause analysis on every P1.
Planning a build, upgrade or migration instead? See AceMQ Solace consulting and services.
Customers




Partners


Incident Triage
Spool, HA, redelivery and event mesh problems on Solace brokers, each paired with the broker statistic or log an engineer reads first.
| Symptom you're seeing | What we check first | Typical resolution |
|---|---|---|
| Publishers rejected or slowed, message spool usage near its quota | Per-queue spool usage and the queue with no active consumer (show message-spool detail, show queue <name> detail); queues with no maximum spool usage set; messages waiting on a consumer that was decommissioned. | Bound each queue's maximum spool usage, remove or drain the orphan queue, add a dead message queue and a TTL where the data allows it. |
| One slow subscriber stalls delivery to others on the same message VPN | Flow state for the client (show client <name> detail), unacknowledged messages against the queue's max-delivered-unacked-msgs-per-flow, the client's window size, and the queue depth behind the flow. | Fix the consumer's acknowledgement pattern, size the window for the round trip, set a spool limit so one consumer cannot fill the broker. |
| HA pair failed over but clients did not reconnect, or the pair will not stay in sync | show redundancy and show config-sync on both nodes and the monitoring node; whether the redundancy link is up; whether the client's reconnect settings match the measured failover time. | Repair the link or the config-sync state, correct reconnect retries and timeouts in the client, rehearse a planned switchover. |
| Messages redelivered in a loop to the same consumer | The queue's max-redelivery setting (zero means retry forever), the consumer's processing errors, and whether a poison message is blocking an exclusive queue. | Set a maximum redelivery count with a dead message queue, quarantine and inspect the poison message, fix the consumer. |
| Event mesh link between two brokers is down or lagging | Dynamic message routing link state and the bridge queue depth between the brokers; certificate or network changes on the inter-broker path; subscription propagation. | Restore the link, correct the TLS chain, validate subscription propagation, check the receiving side's spool. |
| Client connections refused after growth | Message VPN and broker limits for connections and subscriptions (max-connections, max-subscriptions); client profile and ACL profile rules for the new application. | Raise the limit that is actually reached, correct the profile rules, set per-VPN limits so one team cannot consume the broker. |
| Disk or spool filling after a replay or a large backlog | Replay log size and retention, spool pressure while the backlog drains, and whether consumers were scaled for catch-up. | Size replay and spool for the drain rate, throttle the replay, add consumers for catch-up. |
Every P1 comes with a written cause naming the queue, message VPN or link involved.
Deployment coverage
| Deployment | What AceMQ covers | What Solace covers | Excluded |
|---|---|---|---|
| Software event brokers on virtual machines and containers | 24/7 incident response, message spool and queue design, client profile and ACL profile review, HA pair behavior, upgrades, written RCA on P1s | Solace supplies releases and fixes for the software and takes product defect cases (solace.com/support, read 17 November 2026) | A patched Solace binary |
| Appliance event brokers | Configuration, queue and spool behavior, HA pair operation and upgrade rehearsal on the appliance broker | Return Material Authorization and hardware sparing or repair, as its support plan describes | Hardware faults and replacement |
| Event mesh and dynamic message routing between brokers | Link design, subscription propagation checks, inter-broker TLS, bridge queue sizing, failure drills across sites | Solace supports the product features it ships | Network paths between your sites |
| Solace Cloud | The application and configuration layer you control: queues, topics, client behavior, ACLs | Solace operates its cloud service and supports it through its own console and plan | Service-level faults |
| Replication and disaster recovery sites | Replication state checks, failover drills, runbooks that are tested | Solace supports the product | Storage and WAN faults |
| Clients and APIs (Java, JMS, .NET, C, Python, JavaScript, MQTT, AMQP) | Reconnect behavior, acknowledgement patterns, flow control, protocol mapping | Solace supports its own APIs | Application code defects |
Solace and independent support
Solace provides enterprise support for its own event brokers through its support plan, with 24/7/365 hours. Independent providers such as AceMQ support Solace operations and troubleshooting alongside that plan or without it. Solace fixes product defects and supplies hardware; an independent contract adds an engineer who works your configuration and your clients.
Solace's own Advanced Services target is 15 minutes for Severity 1, and ours starts from 15 minutes by plan, so response time alone is not the reason to add an independent contract. The reason is scope. A spool that fills because nobody owns a queue, or a client that reconnects too slowly after a switchover, is configuration, and an engineer who reads your setup is faster than a defect ticket.
Why AceMQ
On the first call
On the first call we ask for four artefacts because they separate causes in minutes.
From the active broker. It shows total spool use against quota and tells us whether one queue or the whole broker is the problem.
For the queue behind the symptom. Depth, consumer count, redelivery settings and the maximum spool size separate an orphan queue from a slow consumer.
On both nodes. They show which node is active, whether the link is up and whether the pair agrees on configuration.
Reconnect retries, the acknowledgement mode and the window size. Most "the broker lost messages" calls end here.
Escalation Path
A message spool at its quota stops publishers in minutes. The engineer on the line starts with queue depth and consumer binds, not with questions about what Solace is.
Your plan sets the clock: a full message spool is treated differently from an upgrade planning query.
Pricing
Solace support is usually priced from $24,000 to $72,000 a year. The number of brokers and message VPNs, appliance versus software, event mesh scale and your response plan shape where you fall.
Prices are in US dollars (USD). Response targets start at 15 minutes by plan, 24/7, and quote requests are answered within 1 hour. A contract that also covers other messaging systems earns a discount.
Support options
AceMQ offers Solace support in four tiers, each handing more of the day-to-day work to AceMQ: 24/7 support and incident management for brokers your team runs, proactive monitoring that catches spool, HA and mesh drift first, and managed Solace, where AceMQ runs your brokers day to day and quotes the scope per estate. Hardware repair and replacement stay with Solace in every tier.
Your messaging team runs Solace, we assist
Best for: Teams operating Solace brokers who want an expert reachable for spool, failover or delivery problems.
Planned reviews of queues, limits and HA health
Best for: Teams that want spool trends, slow consumers and failover behavior examined before they reach a limit.
We operate your Solace brokers in your estate
Best for: Enterprises that keep brokers on their own sites and want AceMQ running them day to day.
AceMQ runs the event mesh for you
Best for: Teams that need guaranteed messaging without building a specialist team to look after the brokers.
One partner for Solace
The team also supports IBM MQ, ActiveMQ and Kafka, and plans moves between messaging platforms.
Solace provides enterprise support for its event brokers through its support plan, with 24/7/365 hours and a 15-minute Severity 1 target on Advanced Services (solace.com/support, read 17 November 2026). Independent providers such as AceMQ support Solace operations, troubleshooting and incident response beside that plan or on its own, with response targets from 15 minutes, by plan.
Yes, and that is the usual set-up. Solace handles product defects, releases and hardware. We handle why a spool filled, why clients did not reconnect after a switchover and how queues, ACLs and mesh links should be configured. When a defect is involved we open the case with Solace and drive it.
We support the configuration and operations of both: queues, spool, HA pairs, upgrades and clients. Hardware replacement and repair stay with Solace, whose support plan requires a Return Material Authorization for it (solace.com/support, read 17 November 2026).
A spool fills when messages arrive faster than consumers acknowledge them, or when a queue has no active consumer. We check queue depth by endpoint, consumer binds and any large replay, then adjust quotas or the slow consumer. Raising the quota alone only delays the problem.
After a failover, clients reconnect only if they list both broker addresses and retry. We check the HA pair's sync state, the client connection URL, reconnect settings and any firewall or DNS pointing at one node, then test the switch in a planned window.
Spool and queue sizing, message VPN limits, client and ACL profile review, HA pair and config-sync state, event mesh link and bridge queue state, certificate expiry, replay and replication settings, client reconnect behavior, and a ranked plan. Quotes are answered within 1 hour.
We start from the release each broker is on, check client and API compatibility, upgrade the backup node first in an HA pair, rehearse a switchover, then upgrade the other node and keep a rollback. The first step is an inventory of broker versions, topology and clients.
Yes. Solace supports AMQP, Kafka, MQTT, REST and WebSocket (solace.com/products/event-broker, read 17 November 2026), which makes phased moves with parallel running practical. We map topics, queues and delivery guarantees to the target, run both sides side by side and keep a rollback. See Solace consulting.
They refer to the same broker product line. Solace's pages use the name Solace Event Broker and still carry the PubSub+ name in places (solace.com/products/event-broker, read 17 November 2026). We support the deployed broker whichever name your team uses.
Yes. Many teams keep their Solace plan and add AceMQ for the gaps, such as spool and HA diagnosis at odd hours, upgrade planning or advice across several messaging platforms. You can also hire us for one project, like a health check, with no annual commitment.
Annual contracts typically run $24,000 to $72,000 (USD). Inputs are broker count, appliance or software form, message VPNs and HA pairs, event mesh links, and your response plan. Describe the estate in a quote request and we respond within 1 hour.
Let's talk
Send us your broker versions, the HA and mesh topology and the queues that carry your critical traffic. We reply to quote requests within 1 hour.
Production incident right now?Email a Solace Architect