The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Manual binary-log application is a controlled replay: inspect the source log, select coordinates that match the replica’s snapshot, pipe the chosen events through mysqlbinlog into the mysql client, then verify replication state and data. The procedure is useful for point-in-time recovery and narrowly scoped repair, but an incorrect position, mismatched snapshot, or second applier running at the same time can permanently diverge the replica.
Antihost offers Linux VPS plans with full root access, unlimited traffic on a 1 Gbps port and a 14-day free trial.
What manual binary-log replay actually does
MySQL stores committed changes as events in binary-log files. mysqlbinlog reads those events and writes an executable representation; the MySQL client then sends that representation to the target server. Oracle’s MySQL 8.4 documentation describes the operation as processing mysqlbinlog output with the mysql client to apply events from the binary log.
As an Amazon Associate I earn from qualifying purchases.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesFor statement-based events, the output is SQL. Row-based events are represented internally as BINLOG statements. Adding --base64-output=DECODE-ROWS --verbose makes row changes readable while inspecting a log, but those display options are not a substitute for the original event representation during replay.
Manual replay is not a replacement for normal replication. It is an intervention in which you choose a file and position (or a bounded range), stop competing appliers, execute the events, and deliberately return the replica to a known replication configuration.
#1 Best Overall
- Easily store and access 2TB to content on the go with the Seagate Portable Drive, a USB external hard drive
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Prepare the replica and the source
Do preparation before opening a binary log. The following conditions determine whether replay can be safe:
- Have a consistent starting snapshot. The replica must contain a dump or physical snapshot whose contents correspond to a known position in the source’s binary log. A log cannot make an arbitrary, stale snapshot consistent.
- Record the matching coordinates. Save the source binary-log file and position (or the relevant GTID set) captured with the snapshot. Those coordinates are the only defensible starting point.
- Confirm source logging. Binary logging must be enabled on the source, and every server in the topology must have a unique, nonzero
server_id. - Stop competing replication. If the target is an already configured replica, stop or isolate its normal receiver and applier threads before applying events manually. Two appliers changing the same data can interleave transactions and create divergence.
- Use an appropriately privileged account. The account executing generated
BINLOGstatements needsBINLOG_ADMIN(or the deprecatedSUPERprivilege). MySQL also documentsREPLICATION_APPLIERtogether with the privileges required by each event as an alternative arrangement. - Preserve evidence. Work in a maintenance window or disposable clone when possible. Keep the original binary logs, the exact command lines, command output, replica status, and error logs so the operation can be audited or diagnosed.
Do not begin by guessing a position from the time shown in a log. First identify the snapshot’s recorded coordinates, then inspect the surrounding event boundaries.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Choose the correct coordinates
File and position
A file-position topology identifies a source log such as binlog.000123 and a byte position within that file. The position attached to a consistent snapshot is the point at which replay should begin. In MySQL’s replication syntax, SOURCE_LOG_POS=4 represents the start of events in a binary-log file; it is a protocol-level starting point, not a universal instruction to start every recovery at 4. Use 4 only when the snapshot and recovery plan specifically require the beginning of that file.
For a bounded repair, define both a start and a stop position. A stop position is an exclusive boundary in the selected file, so inspect the neighboring event groups and transactions before applying it. Never cut through a transaction merely because the numeric position looks convenient.
GTID coordinates
When both source and replica use GTIDs, the safer normal-resumption method is automatic positioning. Configure SOURCE_AUTO_POSITION = 1 rather than supplying file and position. Do not combine auto-positioning with SOURCE_LOG_FILE or SOURCE_LOG_POS; they are different coordinate systems.
Why the snapshot matters
Binary logs contain changes, not the complete data set. Replaying from a point earlier than the snapshot’s matching coordinate can apply rows twice; starting later leaves changes missing. Filtering differences, a partially loaded dump, or a snapshot taken while writes were inconsistent can produce the same symptoms as a bad position, so validate the snapshot provenance before blaming the log.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Inspect the log before executing anything
Inspection should be a separate, read-only step. For one file and a proposed range, run:
Rank #2
- Easily store and access 5TB of content on the go with the Seagate portable drive, a USB external hard Drive
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
mysqlbinlog --base64-output=DECODE-ROWS --verbose
--start-position=START_POS --stop-position=STOP_POS
binlog.000123 | lessReplace START_POS and STOP_POS with the coordinates recorded for your recovery. Review the output for transaction boundaries, database and table names, DDL, session settings, and any event that could be unsafe on the target. The verbose decoded view is for human review; it does not alter the event stream that will be replayed.
Check whether the range crosses files or includes a transaction that starts before the selected start position. If it does, move the start to a deliberate event-group or transaction boundary. A range that begins in the middle of a transaction can fail, omit prerequisite changes, or leave the target only partly updated.
Apply a local binary-log range
One file
After review, apply exactly the approved range:
mysqlbinlog --start-position=START_POS --stop-position=STOP_POS
binlog.000123 | mysql --binary-mode=1 -u root -pThe shell pipeline reads the selected events and passes them directly to the target MySQL server. Enter the target account’s password when prompted. --binary-mode=1 keeps the client from rewriting binary data while consuming generated BINLOG statements.
Several consecutive files
List consecutive files in chronological order. Apply a start position only to the first file and a stop position only to the final file when the recovery range spans more than one file; otherwise process complete intermediate files. For example:
mysqlbinlog --start-position=START_POS
binlog.000123 binlog.000124
--stop-position=STOP_POS | mysql --binary-mode=1 -u root -pBefore running a multi-file command, verify that the files are from the same source and that their sequence is intact. Preserve the exact command and its terminal output. If the client exits with an error, stop and investigate rather than rerunning the entire range automatically; rerunning can duplicate transactions that already committed.
Read logs that are remote or encrypted at rest
If the required files are not local, stream them from the source with --read-from-remote-server (short form -R):
mysqlbinlog --read-from-remote-server
--host=source.example --port=3306
--user=binlog_reader --password
--ssl-mode=VERIFY_IDENTITY
binlog.000123 | mysql -u root -pThe account used by mysqlbinlog should be limited to the source access required for reading logs. Protect the connection with TLS. MySQL recommends VERIFY_CA or VERIFY_IDENTITY rather than REQUIRED when possible, because the stronger modes validate the server certificate instead of merely encrypting the connection.
Recommended Free Tools
Rank #3
- Easily store and access 1TB to content on the go with the Seagate Portable Drive, a USB external hard drive.Specific uses: Personal
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop. Reformatting may be required for Mac
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Remote streaming does not change coordinate discipline: inspect the exact range first, make sure the source is reachable and serving the expected file, and keep the target applier stopped. If logs are encrypted at rest, decrypt them only in a controlled location or use the remote-read path; do not create untracked copies of sensitive logs.
Return a file-position replica to normal operation
After manual replay has completed and validation shows that the target is ready, configure the source coordinates for the next event to fetch. Stop the receiver and applier while changing the configuration, then issue:
CHANGE REPLICATION SOURCE TO
SOURCE_HOST='source_host',
SOURCE_USER='replication_user',
SOURCE_PASSWORD='password',
SOURCE_LOG_FILE='binlog.000123',
SOURCE_LOG_POS=START_POS;
START REPLICA;Use the next valid coordinate, not automatically the command’s original start position. If the manual range ended immediately before an event, the restart position must point to that event. If you intentionally applied through a transaction, set the coordinate after that transaction. A wrong coordinate can replay an event twice or skip part of a transaction.
For a GTID topology, configure SOURCE_AUTO_POSITION = 1 and omit file-position options. The replica then negotiates missing GTIDs with the source. Do not add SOURCE_LOG_FILE or SOURCE_LOG_POS to an auto-position configuration.
Handling a failed transaction or an intentional skip
Skipping is a last-resort data decision, not a generic way to clear an error. First stop both the receiver and applier threads. Capture SHOW REPLICA STATUS, the error number and text, the relevant source and relay logs, and the executed GTID set where applicable.
Use mysqlbinlog to locate the failed event group and the next valid transaction boundary. Decide whether the transaction was already applied, whether it is safe to omit, and whether an equivalent corrective change must be made manually. Then set restart coordinates deliberately and start the replica. Never advance to a visually convenient byte offset inside a transaction.
Skipping can hide a real schema, privilege, filtering, or data-integrity problem. Preserve the pre-skip evidence and record the reason, operator, coordinates, and validation results. If the target’s state is uncertain, restoring a known-good snapshot is safer than repeatedly advancing positions.
Rank #4
- Easily store and access 4TB of content on the go with the Seagate Portable Drive, a USB external hard drive.Specific uses: Personal
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Validate both replication and data
Replay is not complete when the pipe exits successfully. Run SHOW REPLICA STATUS and review:
- Receiver and applier running states after restart.
- Source file and position, relay coordinates, and read/execute gaps.
- Error numbers and messages; an empty current error is not proof that the data matches.
- Executed GTID sets when GTIDs are enabled.
- Any expected delay while the replica catches up.
Then compare the source and replica at the application level. Use known row counts, targeted key checks, or checksums appropriate to the affected tables. A coordinate match can coexist with divergence if the snapshot was wrong, filters differed, or an applier ran concurrently. Conversely, a temporary coordinate lag can be normal while the replica processes the remaining queue.
Keep the source and replica binary logs, status output, error logs, inspection output, and replay command together. This evidence is essential before resetting replication state, rebuilding a replica, or explaining an incident.
Manual replay or restart replication?
| Decision factor | Manual binary-log replay | Restarting configured replication |
|---|---|---|
| Range control | Exact file and position range can be inspected and applied. | Normal threads continue from their configured coordinates or GTID set. |
| Source reachability | Local files can be used offline; remote or encrypted logs require --read-from-remote-server and protected transport. |
Source connectivity and replication credentials must work continuously. |
| Coordinate model | You choose boundaries explicitly and must avoid splitting transactions. | File-position or GTID machinery advances coordinates automatically. |
| Privileges | The replay account needs BINLOG_ADMIN (or deprecated SUPER), or REPLICATION_APPLIER plus event privileges. |
Configured replication users and the replica’s applier privileges must be valid. |
| Transaction risk | High if a range starts, stops, or resumes at the wrong boundary; a second applier must be stopped. | Lower for a healthy topology because the applier follows recorded boundaries, but existing errors still require diagnosis. |
| Auditability | Exact files, commands, output, and reviewed ranges can be preserved. | Status history and server logs provide the primary record. |
| Rollback options | There is no automatic undo for committed events; restore a snapshot or perform a separately designed correction. | There is likewise no automatic undo for already committed replicated changes. |
| Best fit | Point-in-time recovery, offline repair, or a narrowly bounded intervention after normal replication is isolated. | A consistent replica whose source, credentials, filters, and coordinates are already healthy. |
Common failure modes and recovery choices
Permission errors for BINLOG statements
If the target rejects generated BINLOG statements, verify the executing account’s BINLOG_ADMIN privilege (or the deprecated SUPER equivalent). If using REPLICATION_APPLIER, grant the privileges required by the specific events. Do not solve a privilege error by switching to an all-powerful account without documenting the change.
Duplicate-key or already-applied errors
These often indicate that the selected start position is earlier than the snapshot coordinate or that part of the range was already committed before a retry. Stop the pipeline, compare the target’s data and coordinates with the saved status, and identify the last confirmed transaction before choosing a new boundary. Blindly rerunning can compound the mismatch.
Free tools Windows power users keep installed
One-click scans. No signup required.
Unknown table, column, or schema errors
A DDL event may be missing, filtered, or ordered differently on the target. Inspect preceding events and compare schema definitions. Do not skip the failing transaction until you know whether omitting it will leave later row events impossible or semantically incorrect.
Errors after a network interruption
A broken pipe does not prove that no events committed. Capture the target status and error log, determine the last committed transaction, and resume from the next deliberate boundary. Reconnect only after this check.
Replica will not start after manual application
Run SHOW REPLICA STATUS and check source credentials, the configured coordinate model, filters, and the first reported error. File-position settings require SOURCE_LOG_FILE and SOURCE_LOG_POS; GTID settings require auto-positioning. Remove contradictory options rather than alternating between both models.
A concise operational runbook
- Freeze writes or schedule a maintenance window, and stop the replica’s receiver and applier threads.
- Confirm the snapshot, source log files, server IDs, account privileges, and topology coordinate model.
- Record the snapshot-matching file and position or GTID set.
- Inspect the proposed range with decoded row output and verify complete transaction boundaries.
- Apply the reviewed local or remote range with the exact command recorded.
- On any error, stop, preserve evidence, and identify the last confirmed boundary before retrying.
- Validate status, coordinates, GTIDs, and application-level rows or checksums.
- Configure file-position coordinates or GTID auto-positioning for normal replication, then start the replica.
- Recheck status and data after the replica catches up; retain all logs and command output.
Version and safety scope
The commands and privilege names here target the MySQL 8.4 Reference Manual behavior available on October 15, 2026. MySQL syntax and privilege requirements can differ by release and by managed-service policy, so verify them against the exact server version before execution. Manual replay always reexecutes data modifications; it does not provide an undo log. Treat every position change, skipped transaction, and reset of replication state as a change-management event.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
