{"version":"2.1.248","anchor":"very-large-transcripts-are-reassembled-by-paging-instead-of-","canonical_anchor":"very-large-transcripts-are-reassembled-by-paging-instead-of-","heading":"Very large transcripts are reassembled by paging instead of failing","tier":"notice","area":"Sessions","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248\/e\/very-large-transcripts-are-reassembled-by-paging-instead-of-","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248","markdown":"### Very large transcripts are reassembled by paging instead of failing\n\nVery large transcripts now load by paging instead of erroring, on the newer storage backend.\n\n**What**\n\nWhen a session transcript is too big to read in one go, Claude Code now walks it forward record by record and stitches the pieces back together, so a large history loads instead of erroring. This applies only when the newer storage backend (storage v5) is in use.\n\n**Details**\n\n- A failed whole-file read, or a backend refusal reporting `telemetryCode` `\"EFBIG\"`, triggers the paged read, which itself has a byte cap; exceeding it throws an error carrying `{ code: \"EFBIG\" }`.\n- A cursor that fails to advance throws `loadTranscriptFile: the record cursor did not advance`; a successful paged read logs record and byte counts.\n- Tail reads of a session file fall back to a backward record page when the tail read is unsupported, and the tail result gains a `cut` flag meaning the content was truncated.\n- Which flag selects the storage v5 backend is not settled by this build's code.\n\n**Evidence**\n\n`loadTranscriptFile: the paged transcript is too large`\n\n- Area: Sessions\n- Tier: You'll notice\n- Useful: 4\/5\n- Signal: 2\/5"}