{"version":"2.1.291","anchor":"transcript-loading-drops-the-load-exhausted-unreadable-tra","canonical_anchor":"transcript-loading-drops-the-load-exhausted-unreadable-tra","heading":"One kind of failed session load is no longer reported as an unreadable transcript","tier":"internal","area":"Sessions","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.291\/e\/transcript-loading-drops-the-load-exhausted-unreadable-tra","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.291","markdown":"### One kind of failed session load is no longer reported as an unreadable transcript\n\nA session load that fails with one particular error no longer goes through the 'transcript unreadable' path; the general error handling takes it instead\n\n**What**\n\nWhen Claude Code loads the full history of a session and something goes wrong, it decides how to report the problem. One specific kind of error used to be reported as an unreadable transcript with the reason \"load_exhausted\". That branch has been removed.\n\nThe check that decides whether an error counts as an unreadable transcript no longer includes that kind of error either. It now goes to the general error handling.\n\n**Why**\n\nThis changes how some failed session loads are reported. A failure that used to show up as an unreadable transcript, after loading gave up, is now handled like any other error.\n\n- Area: Sessions\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}