Category: Troubleshooting

RESOURCE_SEMAPHORE waits while SQL Server sits far below max server memory: why the grant pool follows target memory and not the cap, the one query that shows it, and how to let a query run with MAX_GRANT_PERCENT while you free memory.

TempDB belongs to the whole instance, so the question is never whether it is full, it is which session filled it. One query names the session, and user objects, internal objects and the version store each end a different way.

The service will not come up and Windows only says 1067. Where the real reason lives, what errors 17113 and 17058 are actually telling you, and what to do when the log says nothing at all.

Msg 3906 means SQL Server considers the target database read-only. Check its state and recovery role before changing it. A log-shipping secondary may be read-only by design.

RESTORE VERIFYONLY returned Msg 3203 on a truncated backup in this HPAI01 test. The header was readable, but the file was incomplete. Direct RESTORE returned Msg 3287 instead.

A symptom router for the slow SQL Server call. One check per step, in the order that finds the fault fastest, each handing off to the post that owns it.

Three incidents that look alike and share nothing: a BACKUP that failed, a RESTORE that failed, and a restore that is still running. This page routes each one to the post that owns it.

A routing page for every out of space incident on SQL Server: which drive, which file, what log_reuse_wait_desc is telling you, and why shrinking comes last.

Which msdb history table is actually holding the space, why one far back cutoff makes sp_delete_backuphistory hang, and the date stepped purge that finishes.

The job outcome row never carries a cause. Read the step row in msdb.dbo.sysjobhistory first, then route the real message to the owner, proxy, permission, path or truncation behind it.