n8n 1.123.58 - Fixes for CSV upload sizing and a dependency upgrade to shore up reliability
n8n released version 1.123.58 (2026-06-18) addressing a data-table sizing bug that failed to account for pending CSV uploads and including an upgrade to the UUID dependency. These are targeted stability and dependency-maintenance fixes that improve upload handling and reduce operational risk for automation pipelines.
n8n's 1.123.58 patch focuses on two pragmatic maintenance items: accounting for pending CSV uploads in the data-table size budget (PR #32494, commit 0dc6ded) and upgrading the `uuid` dependency. The CSV fix corrects how the UI/backend budget estimates storage and rendering resources while CSVs are still uploading, preventing underestimation of table size and the user-experience, performance, or memory anomalies that can follow. The `uuid` upgrade indicates continued dependency hygiene, likely addressing compatibility or security concerns common in transitively imported libraries.
For businesses relying on n8n to orchestrate data ingestion and integration workflows, the CSV size-budget fix is significant. Many enterprise workflows ingest CSVs of varying sizes; incorrect budgeting can cause UI freezes, failed renders, or backend memory pressure during concurrent uploads, in turn affecting SLA-driven processes. The dependency upgrade reduces supply-chain risk: even minor library updates can remediate vulnerabilities or runtime bugs that would otherwise surface in production. Together these changes are about resilience - reducing operational surprises and making predictable behavior more likely under load.
Actionable guidance for leaders: schedule the patch into your standard release window and run focused QA against CSV-heavy workflows and concurrency scenarios. Validate upload progress, table rendering, and memory/CPU metrics during bulk imports. Treat the `uuid` upgrade as a prompt to review dependency policies - ensure your CI includes dependency-scan gates, SBOM generation, and a rollback plan. Finally, communicate the change to downstream teams (data, ops, and integrators) so they can adjust any upload size assumptions or monitoring alerts tied to ingestion throughput.
Original Source
n8n (GitHub)
