Peer review is the primary mechanism through which software engineering (SE) research establishes quality and credibility, but it is also a social institution that shapes whose work and voices define the field. Yet many researchers increasingly experience this process as stressful, uneven, and misaligned with the goals of rigorous and meaningful scholarship. We analyzed open-ended responses from a community survey conducted ahead of ICSE 2026 FOSE to examine what SE researchers perceive as working well in peer review, and bringing joy, as well as what generates stress. Respondents value rigorous evaluation, intellectual stimulation, mentoring, and community support, but point to excessive submission volumes, time pressure, and inconsistent or superficial reviews as major sources of strain. Less experienced researchers and those with ten or fewer submissions in the past three years were about twice as likely as more experienced and higher-volume submitters to report that peer review is not working well. These patterns suggest that current practices disproportionately burden early-career and most vulnerable researchers, highlighting the need to rebalance workload, accountability, and incentives to sustain a healthy SE research community.